Skip to content

P
PaddleOCR

s920243400 / PaddleOCR
与 Fork 源项目一致

Fork自 PaddlePaddle / PaddleOCR

通知 1
Star 1
Fork 0
P
PaddleOCR
提交 f57a6d1b 编写于 作者: A andyjpaddle
浏览文件
操作

Merge branch 'add_rec_sar' of https://github.com/andyjpaddle/PaddleOCR into add_rec_sar

上级 df4a2f6a 8a7d3a29
隐藏空白更改
内联 并排
Showing with 3723 addition and 829 deletion
README.md
浏览文件 @ f57a6d1b
English | [简体中文](README_ch.md)
<p align="center">
<img src="./doc/PaddleOCR_log.png" align="middle" width = "600"/>
<p align="center">
------------------------------------------------------------------------------------------
<p align="left">
<a href="./LICENSE"><img src="https://img.shields.io/badge/license-Apache%202-dfd.svg"></a>
<a href="https://github.com/PaddlePaddle/PaddleOCR/releases"><img src="https://img.shields.io/github/v/release/PaddlePaddle/PaddleOCR?color=ffa"></a>
<a href=""><img src="https://img.shields.io/badge/python-3.7+-aff.svg"></a>
<a href=""><img src="https://img.shields.io/badge/os-linux%2C%20win%2C%20mac-pink.svg"></a>
<a href=""><img src="https://img.shields.io/pypi/format/PaddleOCR?color=c77"></a>
<a href="https://github.com/PaddlePaddle/PaddleOCR/graphs/contributors"><img src="https://img.shields.io/github/contributors/PaddlePaddle/PaddleOCR?color=9ea"></a>
<a href="https://pypi.org/project/PaddleOCR/"><img src="https://img.shields.io/pypi/dm/PaddleOCR?color=9cf"></a>
<a href="https://github.com/PaddlePaddle/PaddleOCR/stargazers"><img src="https://img.shields.io/github/stars/PaddlePaddle/PaddleOCR?color=ccf"></a>
</p>
## Introduction
PaddleOCR aims to create multilingual, awesome, leading, and practical OCR tools that help users train better models and apply them into practice.
## Notice
PaddleOCR supports both dynamic graph and static graph programming paradigm
- Dynamic graph: dygraph branch (default), **supported by paddle 2.0.0 ([installation](./doc/doc_en/installation_en.md))**
- Static graph: develop branch
**Recent updates**
- 2021.1.21 update more than 25+ multilingual recognition models [models list](./doc/doc_en/models_list_en.md), including：English, Chinese, German, French, Japanese，Spanish，Portuguese Russia Arabic and so on. Models for more languages will continue to be updated [Develop Plan](https://github.com/PaddlePaddle/PaddleOCR/issues/1048).
- 2020.12.15 update Data synthesis tool, i.e., [Style-Text](./StyleText/README.md)，easy to synthesize a large number of images which are similar to the target scene image.
- 2020.11.25 Update a new data annotation tool, i.e., [PPOCRLabel](./PPOCRLabel/README.md), which is helpful to improve the labeling efficiency. Moreover, the labeling results can be used in training of the PP-OCR system directly.
- 2020.9.22 Update the PP-OCR technical article, https://arxiv.org/abs/2009.09941
- PaddleOCR R&D team would like to share the released tools with developers, at 20:15 pm on September 8th, [Live Address](https://live.bilibili.com/21689802).
- 2021.9.7 release PaddleOCR v2.3, [PP-OCRv2](#PP-OCRv2) is proposed. The inference speed of PP-OCRv2 is 220% higher than that of PP-OCR server in CPU device. The F-score of PP-OCRv2 is 7% higher than that of PP-OCR mobile.
- 2021.8.3 released PaddleOCR v2.2, add a new structured documents analysis toolkit, i.e., [PP-Structure](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/ppstructure/README.md), support layout analysis and table recognition (One-key to export chart images to Excel files).
- 2021.4.8 release end-to-end text recognition algorithm [PGNet](https://www.aaai.org/AAAI21Papers/AAAI-2885.WangP.pdf) which is published in AAAI 2021. Find tutorial [here](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/pgnet_en.md)；release multi language recognition [models](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md), support more than 80 languages recognition; especically, the performance of [English recognition model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/models_list_en.md#English) is Optimized.
- [more](./doc/doc_en/update_en.md)
## Features
- PPOCR series of high-quality pre-trained models, comparable to commercial effects
- Ultra lightweight ppocr_mobile series models: detection (3.0M) + direction classifier (1.4M) + recognition (5.0M) = 9.4M
- General ppocr_server series models: detection (47.1M) + direction classifier (1.4M) + recognition (94.9M) = 143.4M
- PP-OCR series of high-quality pre-trained models, comparable to commercial effects
- Ultra lightweight PP-OCRv2 series models: detection (3.1M) + direction classifier (1.4M) + recognition 8.5M) = 13.0M
- Ultra lightweight PP-OCR mobile series models: detection (3.0M) + direction classifier (1.4M) + recognition (5.0M) = 9.4M
- General PP-OCR server series models: detection (47.1M) + direction classifier (1.4M) + recognition (94.9M) = 143.4M
- Support Chinese, English, and digit recognition, vertical text recognition, and long text recognition
- Support multi-language recognition: Korean, Japanese, German, French
- Rich toolkits related to the OCR areas
......@@ -64,39 +82,44 @@ Mobile DEMO experience (based on EasyEdge and Paddle-Lite, supports iOS and Andr
<a name="Supported-Chinese-model-list"></a>
## PP-OCR 2.0 series model list（Update on Dec 15）
**Note** : Compared with [models 1.1](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/models_list_en.md), which are trained with static graph programming paradigm, models 2.0 are the dynamic graph trained version and achieve close performance.
## PP-OCR series model list（Update on September 8th）
| Model introduction | Model name | Recommended scene | Detection model | Direction classifier | Recognition model |
| ------------------------------------------------------------ | ---------------------------- | ----------------- | ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ |
| Chinese and English ultra-lightweight OCR model (9.4M) | ch_ppocr_mobile_v2.0_xx | Mobile & server |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
| Chinese and English general OCR model (143.4M) | ch_ppocr_server_v2.0_xx | Server |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_traingit.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
| Chinese and English ultra-lightweight PP-OCRv2 model（11.6M） | ch_ppocrv2_xx |Mobile&Server|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_distill_train.tar)| [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_train.tar)|
| Chinese and English ultra-lightweight PP-OCR model (9.4M) | ch_ppocr_mobile_v2.0_xx | Mobile & server |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
| Chinese and English general PP-OCR model (143.4M) | ch_ppocr_server_v2.0_xx | Server |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_traingit.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
For more model downloads (including multiple languages), please refer to [PP-OCR v2.0 series model downloads](./doc/doc_en/models_list_en.md).
For more model downloads (including multiple languages), please refer to [PP-OCR series model downloads](./doc/doc_en/models_list_en.md).
For a new language request, please refer to [Guideline for new language_requests](#language_requests).
## Tutorials
- [Installation](./doc/doc_en/installation_en.md)
- [Environment Preparation](./doc/doc_en/environment_en.md)
- [Quick Start](./doc/doc_en/quickstart_en.md)
- [Code Structure](./doc/doc_en/tree_en.md)
- Algorithm Introduction
- [Text Detection Algorithm](./doc/doc_en/algorithm_overview_en.md)
- [Text Recognition Algorithm](./doc/doc_en/algorithm_overview_en.md)
- [PP-OCR Pipeline](#PP-OCR-Pipeline)
- Model Training/Evaluation
- [Text Detection](./doc/doc_en/detection_en.md)
- [Text Recognition](./doc/doc_en/recognition_en.md)
- [Direction Classification](./doc/doc_en/angle_class_en.md)
- [Yml Configuration](./doc/doc_en/config_en.md)
- Inference and Deployment
- [Quick Inference Based on PIP](./doc/doc_en/whl_en.md)
- [Python Inference](./doc/doc_en/inference_en.md)
- [C++ Inference](./deploy/cpp_infer/readme_en.md)
- [Serving](./deploy/pdserving/README.md)
- [Mobile](./deploy/lite/readme_en.md)
- [Benchmark](./doc/doc_en/benchmark_en.md)
- [PaddleOCR Overview and Installation](./doc/doc_en/paddleOCR_overview_en.md)
- PP-OCR Industry Landing: from Training to Deployment
- [PP-OCR Model and Configuration](./doc/doc_en/models_and_config_en.md)
- [PP-OCR Model Download](./doc/doc_en/models_list_en.md)
- [Yml Configuration](./doc/doc_en/config_en.md)
- [Python Inference](./doc/doc_en/inference_en.md)
- [PP-OCR Training](./doc/doc_en/training_en.md)
- [Text Detection](./doc/doc_en/detection_en.md)
- [Text Recognition](./doc/doc_en/recognition_en.md)
- [Direction Classification](./doc/doc_en/angle_class_en.md)
- Inference and Deployment
- [Python Inference](./doc/doc_en/inference_en.md)
- [C++ Inference](./deploy/cpp_infer/readme_en.md)
- [Serving](./deploy/pdserving/README.md)
- [Mobile](./deploy/lite/readme_en.md)
- [Benchmark](./doc/doc_en/benchmark_en.md)
- [PP-Structure: Information Extraction](./ppstructure/README.md)
- [Layout Parser](./ppstructure/layout/README.md)
- [Table Recognition](./ppstructure/table/README.md)
- Academic Circles
- [Two-stage Algorithm](./doc/doc_en/algorithm_overview_en.md)
- [PGNet Algorithm](./doc/doc_en/algorithm_overview_en.md)
- Data Annotation and Synthesis
- [Semi-automatic Annotation Tool: PPOCRLabel](./PPOCRLabel/README.md)
- [Data Synthesis Tool: Style-Text](./StyleText/README.md)
......@@ -114,17 +137,18 @@ For a new language request, please refer to [Guideline for new language_requests
- [License](#LICENSE)
- [Contribution](#CONTRIBUTION)
<a name="PP-OCRv2"></a>
## PP-OCRv2 Pipeline
<div align="center">
<img src="./doc/ppocrv2_framework.jpg" width="800">
</div>
<a name="PP-OCR-Pipeline"></a>
[1] PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of three parts: DB text detection, detection frame correction and CRNN text recognition. The system adopts 19 effective strategies from 8 aspects including backbone network selection and adjustment, prediction head design, data augmentation, learning rate transformation strategy, regularization parameter selection, pre-training model use, and automatic model tailoring and quantization to optimize and slim down the models of each module (as shown in the green box above). The final results are an ultra-lightweight Chinese and English OCR model with an overall size of 3.5M and a 2.8M English digital OCR model. For more details, please refer to the PP-OCR technical article (https://arxiv.org/abs/2009.09941).
## PP-OCR Pipeline
[2] On the basis of PP-OCR, PP-OCRv2 is further optimized in five aspects. The detection model adopts CML(Collaborative Mutual Learning) knowledge distillation strategy and CopyPaste data expansion strategy; The recognition model adopts LCNet lightweight backbone network, U-DML knowledge distillation strategy and enhanced CTC loss function improvement (as shown in the red box above), which further improves the inference speed and prediction effect. For more details, please refer to the technical report of PP-OCRv2 (arXiv link is coming soon).
<div align="center">
<img src="./doc/ppocr_framework.png" width="800">
</div>
PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of three parts: DB text detection[2], detection frame correction and CRNN text recognition[7]. The system adopts 19 effective strategies from 8 aspects including backbone network selection and adjustment, prediction head design, data augmentation, learning rate transformation strategy, regularization parameter selection, pre-training model use, and automatic model tailoring and quantization to optimize and slim down the models of each module. The final results are an ultra-lightweight Chinese and English OCR model with an overall size of 3.5M and a 2.8M English digital OCR model. For more details, please refer to the PP-OCR technical article (https://arxiv.org/abs/2009.09941). Besides, The implementation of the FPGM Pruner [8] and PACT quantization [9] is based on [PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim).
## Visualization [more](./doc/doc_en/visualization_en.md)
......
README_ch.md
浏览文件 @ f57a6d1b
[English](README.md) | 简体中文
<p align="center">
<img src="./doc/PaddleOCR_log.png" align="middle" width = "600"/>
<p align="center">
------------------------------------------------------------------------------------------
<p align="left">
<a href="./LICENSE"><img src="https://img.shields.io/badge/license-Apache%202-dfd.svg"></a>
<a href="https://github.com/PaddlePaddle/PaddleOCR/releases"><img src="https://img.shields.io/github/v/release/PaddlePaddle/PaddleOCR?color=ffa"></a>
<a href=""><img src="https://img.shields.io/badge/python-3.7+-aff.svg"></a>
<a href=""><img src="https://img.shields.io/badge/os-linux%2C%20win%2C%20mac-pink.svg"></a>
<a href=""><img src="https://img.shields.io/pypi/format/PaddleOCR?color=c77"></a>
<a href="https://github.com/PaddlePaddle/PaddleOCR/graphs/contributors"><img src="https://img.shields.io/github/contributors/PaddlePaddle/PaddleOCR?color=9ea"></a>
<a href="https://pypi.org/project/PaddleOCR/"><img src="https://img.shields.io/pypi/dm/PaddleOCR?color=9cf"></a>
<a href="https://github.com/PaddlePaddle/PaddleOCR/stargazers"><img src="https://img.shields.io/github/stars/PaddlePaddle/PaddleOCR?color=ccf"></a>
</p>
## 简介
PaddleOCR旨在打造一套丰富、领先、且实用的OCR工具库，助力使用者训练出更好的模型，并应用落地。
## 注意
PaddleOCR同时支持动态图与静态图两种编程范式
- 动态图版本：dygraph分支（默认），需将paddle版本升级至2.0.0（[快速安装](./doc/doc_ch/installation.md)）
- 静态图版本：develop分支
**近期更新**
- 2021.4.8 release 2.1版本，新增AAAI 2021论文[端到端识别算法PGNet](./doc/doc_ch/pgnet.md)开源，[多语言模型](./doc/doc_ch/multi_languages.md)支持种类增加到80+。
- 2021.2.1 [FAQ](./doc/doc_ch/FAQ.md)新增5个高频问题，总数162个，每周一都会更新，欢迎大家持续关注。
- 2021.1.21 更新多语言识别模型，目前支持语种超过27种，包括中文简体、中文繁体、英文、法文、德文、韩文、日文、意大利文、西班牙文、葡萄牙文、俄罗斯文、阿拉伯文等，后续计划可以参考[多语言研发计划](https://github.com/PaddlePaddle/PaddleOCR/issues/1048)
- 2020.12.15 更新数据合成工具[Style-Text](./StyleText/README_ch.md)，可以批量合成大量与目标场景类似的图像，在多个场景验证，效果明显提升。
- 2020.11.25 更新半自动标注工具[PPOCRLabel](./PPOCRLabel/README_ch.md)，辅助开发者高效完成标注任务，输出格式与PP-OCR训练任务完美衔接。
- 2020.9.22 更新PP-OCR技术文章，https://arxiv.org/abs/2009.09941
- [More](./doc/doc_ch/update.md)
- PaddleOCR研发团队对最新发版内容技术深入解读，9月8日晚上20:15，[直播地址](https://live.bilibili.com/21689802)。
- 2021.9.7 发布PaddleOCR v2.3，发布[PP-OCRv2](#PP-OCRv2)，CPU推理速度相比于PP-OCR server提升220%；效果相比于PP-OCR mobile 提升7%。
- 2021.8.3 发布PaddleOCR v2.2，新增文档结构分析[PP-Structure](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/ppstructure/README_ch.md)工具包，支持版面分析与表格识别（含Excel导出）。
- 2021.6.29 [FAQ](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/FAQ.md)新增5个高频问题，总数248个，每周一都会更新，欢迎大家持续关注。
- 2021.4.8 release 2.1版本，新增AAAI 2021论文[端到端识别算法PGNet](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/pgnet.md)开源，[多语言模型](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/multi_languages.md)支持种类增加到80+。
- [More](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/update.md)
## 特性
- PPOCR系列高质量预训练模型，准确的识别效果
- 超轻量ppocr_mobile移动端系列：检测（3.0M）+方向分类器（1.4M）+ 识别（5.0M）= 9.4M
- 通用ppocr_server系列：检测（47.1M）+方向分类器（1.4M）+ 识别（94.9M）= 143.4M
- PP-OCR系列高质量预训练模型，准确的识别效果
- 超轻量PP-OCRv2系列：检测（3.1M）+ 方向分类器（1.4M）+ 识别（8.5M）= 13.0M
- 超轻量PP-OCR mobile移动端系列：检测（3.0M）+方向分类器（1.4M）+ 识别（5.0M）= 9.4M
- 通用PPOCR server系列：检测（47.1M）+方向分类器（1.4M）+ 识别（94.9M）= 143.4M
- 支持中英文数字组合识别、竖排文本识别、长文本识别
- 支持多语言识别：韩语、日语、德语、法语
- 丰富易用的OCR相关工具组件
- 半自动数据标注工具PPOCRLabel：支持快速高效的数据标注
- 数据合成工具Style-Text：批量合成大量与目标场景类似的图像
- 文档分析能力PP-Structure：版面分析与表格识别
- 支持用户自定义训练，提供丰富的预测推理部署方案
- 支持PIP快速安装使用
- 可运行于Linux、Windows、MacOS等多种系统
......@@ -39,7 +54,7 @@ PaddleOCR同时支持动态图与静态图两种编程范式
<img src="doc/imgs_results/ch_ppocr_mobile_v2.0/00018069.jpg" width="800">
</div>
上图是通用ppocr_server模型效果展示，更多效果图请见[效果展示页面](./doc/doc_ch/visualization.md)。
上图是通用PP-OCR server模型效果展示，更多效果图请见[效果展示页面](./doc/doc_ch/visualization.md)。
<a name="欢迎加入PaddleOCR技术交流群"></a>
## 欢迎加入PaddleOCR技术交流群
......@@ -62,71 +77,78 @@ PaddleOCR同时支持动态图与静态图两种编程范式
- 代码体验：从[快速安装](./doc/doc_ch/quickstart.md) 开始
<a name="模型下载"></a>
## PP-OCR 2.0系列模型列表（更新中）
**说明** ：2.0版模型和[1.1版模型](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/models_list.md)的主要区别在于动态图训练vs.静态图训练，模型性能上无明显差距。
## PP-OCR系列模型列表（更新中）
| 模型简介 | 模型名称 |推荐场景 | 检测模型 | 方向分类器 | 识别模型 |
| ------------ | --------------- | ----------------|---- | ---------- | -------- |
| 中英文超轻量OCR模型（9.4M） | ch_ppocr_mobile_v2.0_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
| 中英文通用OCR模型（143.4M） |ch_ppocr_server_v2.0_xx|服务器端 |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
| 中英文超轻量PP-OCRv2模型（11.6M） | ch_ppocrv2_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_distill_train.tar)| [推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_train.tar)|
| 中英文超轻量PP-OCR mobile模型（9.4M） | ch_ppocr_mobile_v2.0_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
| 中英文通用PP-OCR server模型（143.4M） |ch_ppocr_server_v2.0_xx|服务器端 |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
更多模型下载（包括多语言），可以参考[PP-OCR v2.0 系列模型下载](./doc/doc_ch/models_list.md)
更多模型下载（包括多语言），可以参考[PP-OCR 系列模型下载](./doc/doc_ch/models_list.md)
## 文档教程
- [快速安装](./doc/doc_ch/installation.md)
- [中文OCR模型快速使用](./doc/doc_ch/quickstart.md)
- [多语言OCR模型快速使用](./doc/doc_ch/multi_languages.md)
- [代码组织结构](./doc/doc_ch/tree.md)
- 算法介绍
- [文本检测](./doc/doc_ch/algorithm_overview.md)
- [文本识别](./doc/doc_ch/algorithm_overview.md)
- [PP-OCR Pipeline](#PP-OCR)
- [端到端PGNet算法](./doc/doc_ch/pgnet.md)
- 模型训练/评估
- [文本检测](./doc/doc_ch/detection.md)
- [文本识别](./doc/doc_ch/recognition.md)
- [方向分类器](./doc/doc_ch/angle_class.md)
- [yml参数配置文件介绍](./doc/doc_ch/config.md)
- 预测部署
- [基于pip安装whl包快速推理](./doc/doc_ch/whl.md)
- [基于Python脚本预测引擎推理](./doc/doc_ch/inference.md)
- [基于C++预测引擎推理](./deploy/cpp_infer/readme.md)
- [服务化部署](./deploy/pdserving/README_CN.md)
- [端侧部署](./deploy/lite/readme.md)
- [Benchmark](./doc/doc_ch/benchmark.md)
- 数据集
- [通用中英文OCR数据集](./doc/doc_ch/datasets.md)
- [手写中文OCR数据集](./doc/doc_ch/handwritten_datasets.md)
- [垂类多语言OCR数据集](./doc/doc_ch/vertical_and_multilingual_datasets.md)
- [运行环境准备](./doc/doc_ch/environment.md)
- [快速开始（中英文/多语言/文档分析）](./doc/doc_ch/quickstart.md)
- [PaddleOCR全景图与项目克隆](./doc/doc_ch/paddleOCR_overview.md)
- PP-OCR产业落地：从训练到部署
- [PP-OCR模型与配置文件](./doc/doc_ch/models_and_config.md)
- [PP-OCR模型下载](./doc/doc_ch/models_list.md)
- [配置文件内容与生成](./doc/doc_ch/config.md)
- [模型库快速使用](./doc/doc_ch/inference.md)
- [PP-OCR模型训练](./doc/doc_ch/training.md)
- [文本检测](./doc/doc_ch/detection.md)
- [文本识别](./doc/doc_ch/recognition.md)
- [方向分类器](./doc/doc_ch/angle_class.md)
- PP-OCR模型推理部署
- [基于Python脚本预测引擎推理](./doc/doc_ch/inference.md)
- [基于C++预测引擎推理](./deploy/cpp_infer/readme.md)
- [服务化部署](./deploy/pdserving/README_CN.md)
- [端侧部署](./deploy/lite/readme.md)
- [Benchmark](./doc/doc_ch/benchmark.md)
- [PP-Structure信息提取](./ppstructure/README_ch.md)
- [版面分析](./ppstructure/layout/README_ch.md)
- [表格识别](./ppstructure/table/README_ch.md)
- 数据标注与合成
- [半自动标注工具PPOCRLabel](./PPOCRLabel/README_ch.md)
- [数据合成工具Style-Text](./StyleText/README_ch.md)
- [其它数据标注工具](./doc/doc_ch/data_annotation.md)
- [其它数据合成工具](./doc/doc_ch/data_synthesis.md)
- OCR学术圈
- [两阶段模型介绍与下载](./doc/doc_ch/algorithm_overview.md)
- [端到端PGNet算法](./doc/doc_ch/pgnet.md)
- 数据集
- [通用中英文OCR数据集](./doc/doc_ch/datasets.md)
- [手写中文OCR数据集](./doc/doc_ch/handwritten_datasets.md)
- [垂类多语言OCR数据集](./doc/doc_ch/vertical_and_multilingual_datasets.md)
- [效果展示](#效果展示)
- FAQ
- [【精选】OCR精选10个问题](./doc/doc_ch/FAQ.md)
- [【理论篇】OCR通用32个问题](./doc/doc_ch/FAQ.md)
- [【实战篇】PaddleOCR实战110个问题](./doc/doc_ch/FAQ.md)
- [【理论篇】OCR通用50个问题](./doc/doc_ch/FAQ.md)
- [【实战篇】PaddleOCR实战183个问题](./doc/doc_ch/FAQ.md)
- [技术交流群](#欢迎加入PaddleOCR技术交流群)
- [参考文献](./doc/doc_ch/reference.md)
- [许可证书](#许可证书)
- [贡献代码](#贡献代码)
- [代码组织结构](./doc/doc_ch/tree.md)
<a name="PP-OCRv2"></a>
<a name="PP-OCR"></a>
## PP-OCR Pipeline
## PP-OCRv2 Pipeline
<div align="center">
<img src="./doc/ppocr_framework.png" width="800">
<img src="./doc/ppocrv2_framework.jpg" width="800">
</div>
PP-OCR是一个实用的超轻量OCR系统。主要由DB文本检测[2]、检测框矫正和CRNN文本识别三部分组成[7]。该系统从骨干网络选择和调整、预测头部的设计、数据增强、学习率变换策略、正则化参数选择、预训练模型使用以及模型自动裁剪量化8个方面，采用19个有效策略，对各个模块的模型进行效果调优和瘦身，最终得到整体大小为3.5M的超轻量中英文OCR和2.8M的英文数字OCR。更多细节请参考PP-OCR技术方案 https://arxiv.org/abs/2009.09941 。其中FPGM裁剪器[8]和PACT量化[9]的实现可以参考[PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim)。
[1] PP-OCR是一个实用的超轻量OCR系统。主要由DB文本检测、检测框矫正和CRNN文本识别三部分组成。该系统从骨干网络选择和调整、预测头部的设计、数据增强、学习率变换策略、正则化参数选择、预训练模型使用以及模型自动裁剪量化8个方面，采用19个有效策略，对各个模块的模型进行效果调优和瘦身(如绿框所示)，最终得到整体大小为3.5M的超轻量中英文OCR和2.8M的英文数字OCR。更多细节请参考PP-OCR技术方案 https://arxiv.org/abs/2009.09941
[2] PP-OCRv2在PP-OCR的基础上，进一步在5个方面重点优化，检测模型采用CML协同互学习知识蒸馏策略和CopyPaste数据增广策略；识别模型采用LCNet轻量级骨干网络、UDML 改进知识蒸馏策略和Enhanced CTC loss损失函数改进（如上图红框所示），进一步在推理速度和预测效果上取得明显提升。更多细节请参考PP-OCR技术方案（arxiv链接生成中）。
<a name="效果展示"></a>
## 效果展示 [more](./doc/doc_ch/visualization.md)
- 中文模型
<div align="center">
<img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/test_add_91.jpg" width="800">
<img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/00015504.jpg" width="800">
<img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/00056221.jpg" width="800">
<img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/rotate_00052204.jpg" width="800">
</div>
......
configs/det/ch_ppocr_v2.1/ch_det_lite_train_cml_v2.1.yml
浏览文件 @ f57a6d1b
......@@ -8,7 +8,7 @@ Global:
# evaluation is run every 5000 iterations after the 4000th iteration
eval_batch_step: [3000, 2000]
cal_metric_during_train: False
pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
pretrained_model: ./pretrain_models/ch_ppocr_mobile_v2.1_det_distill_train/best_accuracy
checkpoints:
save_inference_dir:
use_visualdl: False
......@@ -19,30 +19,26 @@ Architecture:
name: DistillationModel
algorithm: Distillation
Models:
Student:
pretrained: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
freeze_params: false
Teacher:
freeze_params: true
return_all_feats: false
model_type: det
algorithm: DB
Transform:
Backbone:
name: MobileNetV3
scale: 0.5
model_name: large
disable_se: True
name: ResNet
layers: 18
Neck:
name: DBFPN
out_channels: 96
out_channels: 256
Head:
name: DBHead
k: 50
Student2:
pretrained: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
Student:
freeze_params: false
return_all_feats: false
model_type: det
algorithm: DB
Transform:
Backbone:
name: MobileNetV3
scale: 0.5
......@@ -54,23 +50,24 @@ Architecture:
Head:
name: DBHead
k: 50
Teacher:
pretrained: ./pretrain_models/ch_ppocr_server_v2.0_det_train/best_accuracy
freeze_params: true
Student2:
freeze_params: false
return_all_feats: false
model_type: det
algorithm: DB
Transform:
Backbone:
name: ResNet
layers: 18
name: MobileNetV3
scale: 0.5
model_name: large
disable_se: True
Neck:
name: DBFPN
out_channels: 256
out_channels: 96
Head:
name: DBHead
k: 50
Loss:
name: CombinedLoss
loss_config_list:
......
configs/det/ch_ppocr_v2.1/ch_det_mv3_db_v2.1_student.yml 0 → 100644
浏览文件 @ f57a6d1b
Global:
use_gpu: true
epoch_num: 1200
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/ch_db_mv3/
save_epoch_step: 1200
# evaluation is run every 5000 iterations after the 4000th iteration
eval_batch_step: [0, 400]
cal_metric_during_train: False
pretrained_model: ./pretrain_models/student.pdparams
checkpoints:
save_inference_dir:
use_visualdl: False
infer_img: doc/imgs_en/img_10.jpg
save_res_path: ./output/det_db/predicts_db.txt
Architecture:
model_type: det
algorithm: DB
Transform:
Backbone:
name: MobileNetV3
scale: 0.5
model_name: large
disable_se: True
Neck:
name: DBFPN
out_channels: 96
Head:
name: DBHead
k: 50
Loss:
name: DBLoss
balance_loss: true
main_loss_type: DiceLoss
alpha: 5
beta: 10
ohem_ratio: 3
Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.001
warmup_epoch: 2
regularizer:
name: 'L2'
factor: 0
PostProcess:
name: DBPostProcess
thresh: 0.3
box_thresh: 0.6
max_candidates: 1000
unclip_ratio: 1.5
Metric:
name: DetMetric
main_indicator: hmean
Train:
dataset:
name: SimpleDataSet
data_dir: ./train_data/icdar2015/text_localization/
label_file_list:
- ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
ratio_list: [1.0]
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- DetLabelEncode: # Class handling label
- IaaAugment:
augmenter_args:
- { 'type': Fliplr, 'args': { 'p': 0.5 } }
- { 'type': Affine, 'args': { 'rotate': [-10, 10] } }
- { 'type': Resize, 'args': { 'size': [0.5, 3] } }
- EastRandomCropData:
size: [960, 960]
max_tries: 50
keep_ratio: true
- MakeBorderMap:
shrink_ratio: 0.4
thresh_min: 0.3
thresh_max: 0.7
- MakeShrinkMap:
shrink_ratio: 0.4
min_text_size: 8
- NormalizeImage:
scale: 1./255.
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: 'hwc'
- ToCHWImage:
- KeepKeys:
keep_keys: ['image', 'threshold_map', 'threshold_mask', 'shrink_map', 'shrink_mask'] # the order of the dataloader list
loader:
shuffle: True
drop_last: False
batch_size_per_card: 8
num_workers: 4
Eval:
dataset:
name: SimpleDataSet
data_dir: ./train_data/icdar2015/text_localization/
label_file_list:
- ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- DetLabelEncode: # Class handling label
- DetResizeForTest:
# image_shape: [736, 1280]
- NormalizeImage:
scale: 1./255.
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: 'hwc'
- ToCHWImage:
- KeepKeys:
keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
loader:
shuffle: False
drop_last: False
batch_size_per_card: 1 # must be 1
num_workers: 2
configs/rec/rec_icdar15_train.yml
浏览文件 @ f57a6d1b
......@@ -14,8 +14,8 @@ Global:
use_visualdl: False
infer_img: doc/imgs_words_en/word_10.png
# for data or label process
character_dict_path: ppocr/utils/ic15_dict.txt
character_type: ch
character_dict_path: ppocr/utils/en_dict.txt
character_type: EN
max_text_length: 25
infer_mode: False
use_space_char: False
......
deploy/cpp_infer/CMakeLists.txt
浏览文件 @ f57a6d1b
project(ppocr CXX C)
cmake_minimum_required(VERSION 3.14)
option(WITH_MKL "Compile demo with MKL/OpenBlas support, default use MKL." ON)
option(WITH_GPU "Compile demo with GPU/CPU, default use CPU." OFF)
......@@ -206,13 +207,12 @@ endif()
set(DEPS ${DEPS} ${OpenCV_LIBS})
include(ExternalProject)
include(FetchContent)
include(external-cmake/auto-log.cmake)
include_directories(${CMAKE_CURRENT_BINARY_DIR}/autolog/src/extern_Autolog/auto_log)
include_directories(${FETCHCONTENT_BASE_DIR}/extern_autolog-src)
AUX_SOURCE_DIRECTORY(./src SRCS)
add_executable(${DEMO_NAME} ${SRCS})
target_link_libraries(${DEMO_NAME} ${DEPS})
if (WIN32 AND WITH_MKL)
......
deploy/cpp_infer/docs/vs2019_build_withgpu_config.png 0 → 100644
浏览文件 @ f57a6d1b
因为 它太大了无法显示 image diff 。你可以改为 查看blob。
deploy/cpp_infer/docs/windows_vs2019_build.md
浏览文件 @ f57a6d1b
......@@ -5,20 +5,20 @@ PaddleOCR在Windows 平台下基于`Visual Studio 2019 Community` 进行了测
## 前置条件
* Visual Studio 2019
* CUDA 9.0 / CUDA 10.0，cudnn 7+ （仅在使用GPU版本的预测库时需要）
* CUDA 10.2，cudnn 7+ （仅在使用GPU版本的预测库时需要）
* CMake 3.0+
请确保系统已经安装好上述基本软件，我们使用的是`VS2019`的社区版。
**下面所有示例以工作目录为 `D:\projects`演示**。
### Step1: 下载PaddlePaddle C++ 预测库 fluid_inference
### Step1: 下载PaddlePaddle C++ 预测库 paddle_inference
PaddlePaddle C++ 预测库针对不同的`CPU`和`CUDA`版本提供了不同的预编译版本，请根据实际情况下载: [C++预测库下载列表](https://paddleinference.paddlepaddle.org.cn/user_guides/download_lib.html#windows)
解压后`D:\projects\fluid_inference`目录包含内容为：
解压后`D:\projects\paddle_inference`目录包含内容为：
```
fluid_inference
paddle_inference
├── paddle # paddle核心库和头文件
|
├── third_party # 第三方依赖库和头文件
......@@ -46,13 +46,13 @@ fluid_inference
![step2.2](https://paddleseg.bj.bcebos.com/inference/vs2019_step3.png)
3. 点击：`项目`->`cpp_inference_demo的CMake设置`
3. 点击：`项目`->`CMake设置`
![step3](https://paddleseg.bj.bcebos.com/inference/vs2019_step4.png)
4. 点击`浏览`，分别设置编译选项指定`CUDA`、`CUDNN_LIB`、`OpenCV`、`Paddle预测库`的路径
4. 分别设置编译选项指定`CUDA`、`CUDNN_LIB`、`OpenCV`、`Paddle预测库`的路径
三个编译参数的含义说明如下（带`*`表示仅在使用**GPU版本**预测库时指定, 其中CUDA库版本尽量对齐，**使用9.0、10.0版本，不使用9.2、10.1等版本CUDA库**）：
三个编译参数的含义说明如下（带`*`表示仅在使用**GPU版本**预测库时指定, 其中CUDA库版本尽量对齐）：
| 参数名 | 含义 |
| ---- | ---- |
......@@ -67,6 +67,11 @@ fluid_inference
![step4](https://paddleseg.bj.bcebos.com/inference/vs2019_step5.png)
下面给出with GPU的配置示例：
![step5](./vs2019_build_withgpu_config.png)
**注意：**
CMAKE_BACKWARDS的版本要根据平台安装cmake的版本进行设置。
**设置完成后**, 点击上图中`保存并生成CMake缓存以加载变量`。
5. 点击`生成`->`全部生成`
......@@ -74,24 +79,34 @@ fluid_inference
![step6](https://paddleseg.bj.bcebos.com/inference/vs2019_step6.png)
### Step4: 预测及可视化
### Step4: 预测
上述`Visual Studio 2019`编译产出的可执行文件在`out\build\x64-Release`目录下，打开`cmd`，并切换到该目录：
上述`Visual Studio 2019`编译产出的可执行文件在`out\build\x64-Release\Release`目录下，打开`cmd`，并切换到`D:\projects\PaddleOCR\deploy\cpp_infer\`：
```
cd D:\projects\PaddleOCR\deploy\cpp_infer\out\build\x64-Release
cd D:\projects\PaddleOCR\deploy\cpp_infer
```
可执行文件`ocr_system.exe`即为样例的预测程序，其主要使用方法如下
可执行文件`ppocr.exe`即为样例的预测程序，其主要使用方法如下，更多使用方法可以参考[说明文档](../readme.md)`运行demo`部分。
```shell
#预测图片 `D:\projects\PaddleOCR\doc\imgs\10.jpg`
.\ocr_system.exe D:\projects\PaddleOCR\deploy\cpp_infer\tools\config.txt D:\projects\PaddleOCR\doc\imgs\10.jpg
#识别中文图片 `D:\projects\PaddleOCR\doc\imgs_words\ch\`
.\out\build\x64-Release\Release\ppocr.exe rec --rec_model_dir=D:\projects\PaddleOCR\ch_ppocr_mobile_v2.0_rec_infer --image_dir=D:\projects\PaddleOCR\doc\imgs_words\ch\
#识别英文图片 'D:\projects\PaddleOCR\doc\imgs_words\en\'
.\out\build\x64-Release\Release\ppocr.exe rec --rec_model_dir=D:\projects\PaddleOCR\inference\rec_mv3crnn --image_dir=D:\projects\PaddleOCR\doc\imgs_words\en\ --char_list_file=D:\projects\PaddleOCR\ppocr\utils\dict\en_dict.txt
```
第一个参数为配置文件路径，第二个参数为需要预测的图片路径。
第一个参数为配置文件路径，第二个参数为需要预测的图片路径，第三个参数为配置文本识别的字典。
### 注意
### FQA
* 在Windows下的终端中执行文件exe时，可能会发生乱码的现象，此时需要在终端中输入`CHCP 65001`，将终端的编码方式由GBK编码(默认)改为UTF-8编码，更加具体的解释可以参考这篇博客：[https://blog.csdn.net/qq_35038153/article/details/78430359](https://blog.csdn.net/qq_35038153/article/details/78430359)。
* 编译时，如果报错`错误：C1083 无法打开包括文件:"dirent.h":No such file or directory`，可以参考该[文档](https://blog.csdn.net/Dora_blank/article/details/117740837#41_C1083_direnthNo_such_file_or_directory_54)，新建`dirent.h`文件，并添加到`VC++`的包含目录中。
* 编译时，如果报错`错误：C1083 无法打开包括文件:"dirent.h":No such file or directory`，可以参考该[文档](https://blog.csdn.net/Dora_blank/article/details/117740837#41_C1083_direnthNo_such_file_or_directory_54)，新建`dirent.h`文件，并添加到`utility.cpp`的头文件引用中。同时修改`utility.cpp`70行：`lstat`改成`stat`。
* 编译时，如果报错`Autolog未定义`，新建`autolog.h`文件，内容为：[autolog.h](https://github.com/LDOUBLEV/AutoLog/blob/main/auto_log/autolog.h)，并添加到`main.cpp`的头文件引用中，再次编译。
* 运行时，如果弹窗报错找不到`paddle_inference.dll`或者`openblas.dll`，在`D:\projects\paddle_inference`预测库内找到这两个文件，复制到`D:\projects\PaddleOCR\deploy\cpp_infer\out\build\x64-Release\Release`目录下。不用重新编译，再次运行即可。
* 运行时，弹窗报错提示`应用程序无法正常启动(0xc0000142)`，并且`cmd`窗口内提示`You are using Paddle compiled with TensorRT, but TensorRT dynamic library is not found.`，把tensort目录下的lib里面的所有dll文件复制到release目录下，再次运行即可。
deploy/cpp_infer/external-cmake/auto-log.cmake
浏览文件 @ f57a6d1b
find_package(Git REQUIRED)
message("${CMAKE_BUILD_TYPE}")
include(FetchContent)
set(AUTOLOG_REPOSITORY https://github.com/LDOUBLEV/AutoLog.git)
SET(AUTOLOG_INSTALL_DIR ${CMAKE_CURRENT_BINARY_DIR}/install/Autolog)
set(FETCHCONTENT_BASE_DIR "${CMAKE_CURRENT_BINARY_DIR}/third-party")
ExternalProject_Add(
extern_Autolog
PREFIX autolog
GIT_REPOSITORY ${AUTOLOG_REPOSITORY}
GIT_TAG main
DOWNLOAD_NO_EXTRACT True
INSTALL_COMMAND cmake -E echo "Skipping install step."
FetchContent_Declare(
extern_Autolog
PREFIX autolog
GIT_REPOSITORY https://github.com/LDOUBLEV/AutoLog.git
GIT_TAG main
)
FetchContent_MakeAvailable(extern_Autolog)
deploy/cpp_infer/src/main.cpp
浏览文件 @ f57a6d1b
......@@ -35,6 +35,7 @@
#include <sys/stat.h>
#include <gflags/gflags.h>
#include "auto_log/autolog.h"
DEFINE_bool(use_gpu, false, "Infering with GPU or CPU.");
DEFINE_int32(gpu_id, 0, "Device id of GPU to execute.");
......
doc/doc_ch/FAQ.md
浏览文件 @ f57a6d1b
......@@ -9,38 +9,42 @@
## PaddleOCR常见问题汇总(持续更新)
* [近期更新（2021.2.1）](#近期更新)
* [近期更新（2021.6.29）](#近期更新)
* [【精选】OCR精选10个问题](#OCR精选10个问题)
* [【理论篇】OCR通用32个问题](#OCR通用问题)
* [基础知识7题](#基础知识)
* [数据集7题](#数据集2)
* [模型训练调优18题](#模型训练调优2)
* [【实战篇】PaddleOCR实战120个问题](#PaddleOCR实战问题)
* [使用咨询38题](#使用咨询)
* [数据集18题](#数据集3)
* [模型训练调优30题](#模型训练调优3)
* [预测部署34题](#预测部署3)
* [【理论篇】OCR通用51个问题](#OCR通用问题)
* [基础知识16题](#基础知识)
* [数据集10题](#数据集2)
* [模型训练调优25题](#模型训练调优2)
* [【实战篇】PaddleOCR实战187个问题](#PaddleOCR实战问题)
* [使用咨询80题](#使用咨询)
* [数据集19题](#数据集3)
* [模型训练调优39题](#模型训练调优3)
* [预测部署49题](#预测部署3)
<a name="近期更新"></a>
## 近期更新（2021.2.1）
## 近期更新（2021.6.29）
#### Q3.2.18: PaddleOCR动态图版本如何finetune？
**A**：finetune需要将配置文件里的 Global.load_static_weights设置为false，如果没有此字段可以手动添加，然后将模型地址放到Global.pretrained_model字段下即可。
#### Q2.3.25: 图像正常识别出来的文字是OK的，旋转90度后识别出来的结果比较差，有什么方法可以优化？
A: 整图旋转90之后效果变差是有可能的，因为目前PPOCR默认输入的图片是正向的； 可以自己训练一个整图的方向分类器，放在预测的最前端（可以参照现有方向分类器的方式），或者可以基于规则做一些预处理，比如判断长宽等等。
#### Q3.1.78: 在线demo支持阿拉伯语吗
**A**： 在线demo目前只支持中英文， 多语言的都需要通过whl包自行处理
#### Q3.3.29: 微调v1.1预训练的模型，可以直接用文字垂直排列和上下颠倒的图片吗？还是必须要水平排列的？
**A**：1.1和2.0的模型一样，微调时，垂直排列的文字需要逆时针旋转 90° 后加入训练，上下颠倒的需要旋转为水平的。
#### Q3.1.79: 某个类别的样本比较少，通过增加训练的迭代次数或者是epoch，变相增加小样本的数目，这样能缓解这个问题么？
**A**： 尽量保证类别均衡， 某些类别样本少，可以通过补充合成数据的方式处理；实验证明训练集中出现频次较少的字符，识别效果会比较差，增加迭代次数不能改变样本量少的问题。
#### Q3.3.30: 模型训练过程中如何得到 best_accuracy 模型？
**A**：配置文件里的eval_batch_step字段用来控制多少次iter进行一次eval，在eval完成后会自动生成 best_accuracy 模型，所以如果希望很快就能拿到best_accuracy模型，可以将eval_batch_step改小一点(例如，10)。
#### Q3.1.80: 想把简历上的文字识别出来后，能够把关系一一对应起来，比如姓名和它后面的名字组成一对，籍贯、邮箱、学历等等都和各自的内容关联起来，这个应该如何处理，PPOCR目前支持吗？
**A**: 这样的需求在企业应用中确实比较常见，但往往都是个性化的需求，没有非常规整统一的处理方式。常见的处理方式有如下两种：
1. 对于单一版式、或者版式差异不大的应用场景，可以基于识别场景的一些先验信息，将识别内容进行配对； 比如运用表单结构信息：常见表单"姓名"关键字的后面，往往紧跟的就是名字信息
2. 对于版式多样，或者无固定版式的场景， 需要借助于NLP中的NER技术，给识别内容中的某些字段，赋予key值
#### Q3.4.33: 如何多进程运行paddleocr？
**A**：实例化多个paddleocr服务，然后将服务注册到注册中心，之后通过注册中心统一调度即可，关于注册中心，可以搜索eureka了解一下具体使用，其他的注册中心也行。
由于这部分需求和业务场景强相关，难以用一个统一的模型去处理，目前PPOCR暂不支持。 如果需要用到NER技术，可以参照Paddle团队的另一个开源套件: https://github.com/PaddlePaddle/ERNIE， 其提供的预训练模型ERNIE, 可以帮助提升NER任务的准确率。
#### Q3.4.34: 2.0训练出来的模型，能否在1.1版本上进行部署？
**A**：这个是不建议的，2.0训练出来的模型建议使用dygraph分支里提供的部署代码。
#### Q3.4.49: 同一个模型，c++部署和python部署方式，出来的结果不一致，如何定位？
**A**：有如下几个Debug经验：
1. 优先对一下几个阈值参数是否一致；
2. 排查一下c++代码和python代码的预处理和后处理方式是否一致；
3. 用python在模型输入输出各保存一下二进制文件，排除inference的差异性
<a name="OCR精选10个问题"></a>
## 【精选】OCR精选10个问题
......@@ -76,8 +80,7 @@
**A**：（1）在人眼确认可识别的条件下，对于背景有干扰的文字，首先要保证检测框足够准确，如果检测框不准确，需要考虑是否可以通过过滤颜色等方式对图像预处理并且增加更多相关的训练数据；在识别的部分，注意在训练数据中加入背景干扰类的扩增图像。
（2）如果MobileNet模型不能满足需求，可以尝试ResNet系列大模型来获得更好的效果
。
（2）如果MobileNet模型不能满足需求，可以尝试ResNet系列大模型来获得更好的效果。
#### Q1.1.6：OCR领域常用的评估指标是什么？
......@@ -125,7 +128,7 @@
#### Q1.1.10：PaddleOCR中，对于模型预测加速，CPU加速的途径有哪些？基于TenorRT加速GPU对输入有什么要求？
**A**：（1）CPU可以使用mkldnn进行加速；对于python inference的话，可以把enable_mkldnn改为true，[参考代码](https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/tools/infer/utility.py#L84)，对于cpp inference的话，在配置文件里面配置use_mkldnn 1即可，[参考代码](https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/deploy/cpp_infer/tools/config.txt#L6)
**A**：（1）CPU可以使用mkldnn进行加速；对于python inference的话，可以把enable_mkldnn改为true，[参考代码](https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/tools/infer/utility.py#L99)，对于cpp inference的话，在配置文件里面配置use_mkldnn 1即可，[参考代码](https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/deploy/cpp_infer/tools/config.txt#L6)
（2）GPU需要注意变长输入问题等，TRT6 之后才支持变长输入
......@@ -161,6 +164,39 @@
**A**：处理字符的时候，把多字符的当作一个字就行，字典中每行是一个字。
#### Q2.1.8: 端到端的场景文本识别方法大概分为几种？
**A**：端到端的场景文本识别方法大概分为2种：基于二阶段的方法和基于字符级别的方法。基于两阶段的方法一般先检测文本块，然后提取文本块中的特征用于识别，例如ABCNet；基于字符级别方法直接进行字符检测与识别，直接输出单词的文本框，字符框以及对应的字符类别，例如CharNet。
#### Q2.1.9: 二阶段的端到端的场景文本识别方法的不足有哪些？
**A**: 这类方法一般需要设计针对ROI提取特征的方法，而ROI操作一般比较耗时。
#### Q2.1.10: 基于字符级别的端到端的场景文本识别方法的不足有哪些？
**A**: 这类方法一方面训练时需要加入字符级别的数据，一般使用合成数据，但是合成数据和真实数据有分布Gap。另一方面，现有工作大多数假设文本阅读方向，从上到下，从左到右，没有解决文本方向预测问题。
#### Q2.1.11: AAAI 2021最新的端到端场景文本识别PGNet算法有什么特点？
**A**: PGNet不需要字符级别的标注，NMS操作以及ROI操作。同时提出预测文本行内的阅读顺序模块和基于图的修正模块来提升文本识别效果。该算法是百度自研，近期会在PaddleOCR开源。
#### Q2.1.12: PubTabNet 数据集关注的是什么问题？
**A**: PubTabNet是IBM提出的基于图片格式的表格识别数据集，包含 56.8 万张表格数据的图像，以及图像对应的 html 格式的注释。该数据集的发布推动了表格结构化算法的研发和落地应用。
#### Q2.1.13: PaddleOCR提供的文本识别算法包括哪些？
**A**: PaddleOCR主要提供五种文本识别算法，包括CRNN\StarNet\RARE\Rosetta和SRN, 其中CRNN\StarNet和Rosetta是基于ctc的文字识别算法，RARE是基于attention的文字识别算法；SRN为百度自研的文本识别算法，引入了语义信息，显著提升了准确率。 详情可参照如下页面: [文本识别算法](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.0/doc/doc_ch/algorithm_overview.md#%E6%96%87%E6%9C%AC%E8%AF%86%E5%88%AB%E7%AE%97%E6%B3%95)
#### Q2.1.14: 在识别模型中，为什么降采样残差结构的stride为(2, 1)？
**A**: stride为(2, 1)，表示在图像y方向（高度方向）上stride为2，x方向（宽度方向）上为1。由于待识别的文本图像通常为长方形，这样只在高度方向做下采样，尽量保留宽度方向的序列信息，避免宽度方向下采样后丢失过多的文字信息。
#### Q2.1.15: 文本识别方法CRNN关键技术有哪些？
**A**: CRNN 关键技术包括三部分。（1）CNN提取图像卷积特征。（2）深层双向LSTM网络，在卷积特征的基础上继续提取文字序列特征。（3）Connectionist Temporal Classification(CTC)，解决训练时字符无法对齐的问题。
#### Q2.1.16: 百度自研的SRN文本识别方法特点有哪些？
**A**: SRN文本识别方法特点主要有四个部分：（1）使用Transformer Units（TUs）模块加强图像卷积特征的表达能力。（2）提出Parallel Visual Attention Module（PVAM）模块挖掘特征之间的相互关系。（3）提出Global Semantic Reasoning Module（GSRM）模块挖掘识别结果语义相互关系。（4）提出Visual-Semantic Fusion Decoder（VSFD）模块有效融合PVAM提取的视觉特征和GSRM提取的语义特征。
<a name="数据集2"></a>
### 数据集
......@@ -192,6 +228,16 @@
**A**：SRNet是借鉴GAN中图像到图像转换、风格迁移的想法合成文本数据。不同于通用GAN的方法只选择一个分支，SRNet将文本合成任务分解为三个简单的子模块，提升合成数据的效果。这三个子模块为不带背景的文本风格迁移模块、背景抽取模块和融合模块。PaddleOCR计划将在2020年12月中旬开源基于SRNet的实用模型。
#### Q2.2.8: DBNet如果想使用多边形作为输入，数据标签格式应该如何设定？
**A**：如果想使用多边形作为DBNet的输入，数据标签也应该用多边形来表示。这样子可以更好得拟合弯曲文本。PPOCRLabel暂时只支持矩形框标注和四边形框标注。
#### Q2.2.9: 端到端算法PGNet使用的是什么类型的数据集呢？
**A**: PGNet目前可以使用四点标注数据集，也可以使用多点标注数据集（十四点），多点标注训练的效果要比四点的好，一种可以尝试的策略是先在四点数据集上训练，之后用多点数据集在此基础上继续训练。
#### Q2.2.10: 文档版面分析常用数据集有哪些？
**A**: 文档版面分析常用数据集常用数据集有PubLayNet、TableBank word、TableBank latex等。
<a name="模型训练调优2"></a>
### 模型训练调优
......@@ -254,7 +300,7 @@
**A**：建议可以先了解OCR方向的基础知识，大概了解基础的检测和识别模型算法。然后在Github上可以查看OCR方向相关的repo。目前来看，从内容的完备性来看，PaddleOCR的中英文双语教程文档是有明显优势的，在数据集、模型训练、预测部署文档详实，可以快速入手。而且还有微信用户群答疑，非常适合学习实践。项目地址：[PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)
#### Q3.12：如何识别带空格的英文行文本图像？
#### Q2.3.12：如何识别带空格的英文行文本图像？
**A**：空格识别可以考虑以下两种方案：
......@@ -286,6 +332,33 @@
**A**：SE模块是MobileNetV3网络一个重要模块，目的是估计特征图每个特征通道重要性，给特征图每个特征分配权重，提高网络的表达能力。但是，对于文本检测，输入网络的分辨率比较大，一般是640\*640，利用SE模块估计特征图每个特征通道重要性比较困难，网络提升能力有限，但是该模块又比较耗时，因此在PP-OCR系统中，文本检测的骨干网络没有使用SE模块。实验也表明，当去掉SE模块，超轻量模型大小可以减小40%，文本检测效果基本不受影响。详细可以参考PP-OCR技术文章，https://arxiv.org/abs/2009.09941.
#### Q2.3.19: 参照文档做实际项目时，是重新训练还是在官方训练的基础上进行训练？具体如何操作？
**A**： 基于官方提供的模型，进行finetune的话，收敛会更快一些。 具体操作上，以识别模型训练为例：如果修改了字符文件，可以设置pretraind_model为官方提供的预训练模型
#### Q2.3.20: 如何根据不同的硬件平台选用不同的backbone？
**A**：在不同的硬件上，不同的backbone的速度优势不同，可以根据不同平台的速度-精度图来确定backbone，这里可以参考[PaddleClas模型速度-精度图](https://github.com/PaddlePaddle/PaddleClas/tree/release/2.0/docs/zh_CN/models)。
#### Q2.3.21: 端到端算法PGNet是否支持中文识别，速度会很慢嘛？
**A**：目前开源的PGNet算法模型主要是用于检测英文数字，对于中文的识别需要自己训练，大家可以使用开源的端到端中文数据集，而对于复杂文本（弯曲文本）的识别，也可以自己构造一批数据集针对进行训练，对于推理速度，可以先将模型转换为inference再进行预测，速度应该会相当可观。
#### Q2.3.22: 目前知识蒸馏有哪些主要的实践思路？
**A**：知识蒸馏即利用教师模型指导学生模型的训练，目前有3种主要的蒸馏思路：
1. 基于输出结果的蒸馏，即让学生模型学习教师模型的软标签（分类或者OCR识别等任务中）或者概率热度图（分割等任务中）。
2. 基于特征图的蒸馏，即让学生模型学习教师模型中间层的特征图，拟合中间层的一些特征。
3. 基于关系的蒸馏，针对不同的样本（假设个数为N），教师模型会有不同的输出，那么可以基于不同样本的输出，计算一个NxN的相关性矩阵，可以让学生模型去学习教师模型关于不同样本的相关性矩阵。
当然，知识蒸馏方法日新月异，也欢迎大家提出更多的总结与建议。
#### Q2.3.23: 文档版面分析常用方法有哪些？
**A**: 文档版面分析通常使用通用目标检测方法，包括Faster RCNN系列，YOLO系列等。面向产业实践，建议使用PaddleDetection中精度和效率出色的PP-YOLO v2目标检测方法进行训练。
#### Q2.3.24: 如何识别招牌或者广告图中的艺术字？
**A**: 招牌或者广告图中的艺术字是文本识别一个非常有挑战性的难题，因为艺术字中的单字和印刷体相比，变化非常大。如果需要识别的艺术字是在一个词典列表内，可以将改每个词典认为是一个待识别图像模板，通过通用图像检索识别系统解决识别问题。可以尝试使用PaddleClas的图像识别系统。
#### Q2.3.25: 图像正常识别出来的文字是OK的，旋转90度后识别出来的结果就比较差，有什么方法可以优化？
**A**: 整图旋转90之后效果变差是有可能的，因为目前PPOCR默认输入的图片是正向的； 可以自己训练一个整图的方向分类器，放在预测的最前端（可以参照现有方向分类器的方式），或者可以基于规则做一些预处理，比如判断长宽等等。
<a name="PaddleOCR实战问题"></a>
## 【实战篇】PaddleOCR实战问题
......@@ -361,13 +434,13 @@
（2）inference模型下载时，如果没有安装wget，可直接点击模型链接或将链接地址复制到浏览器进行下载，并解压放置到相应目录。
#### Q3.1.17：PaddleOCR开源的超轻量模型和通用OCR模型的区别？
**A**：目前PaddleOCR开源了2个中文模型，分别是9.4M超轻量中文模型和通用中文OCR模型。两者对比信息如下：
**A**：目前PaddleOCR开源了2个中文模型，分别是8.6M超轻量中文模型和通用中文OCR模型。两者对比信息如下：
- 相同点：两者使用相同的**算法**和**训练数据**；
- 不同点：不同之处在于**骨干网络**和**通道参数**，超轻量模型使用MobileNetV3作为骨干网络，通用模型使用Resnet50_vd作为检测模型backbone，Resnet34_vd作为识别模型backbone，具体参数差异可对比两种模型训练的配置文件.
|模型|骨干网络|检测训练配置|识别训练配置|
|-|-|-|-|
|9.4M超轻量中文OCR模型|MobileNetV3+MobileNetV3|det_mv3_db.yml|rec_chinese_lite_train.yml|
|8.6M超轻量中文OCR模型|MobileNetV3+MobileNetV3|det_mv3_db.yml|rec_chinese_lite_train.yml|
|通用中文OCR模型|Resnet50_vd+Resnet34_vd|det_r50_vd_db.yml|rec_chinese_common_train.yml|
#### Q3.1.18：如何加入自己的检测算法？
......@@ -482,7 +555,239 @@ StyleText的用途主要是：提取style_image中的字体、背景等style信
**A**：Paddle版本问题，请安装2.0版本Paddle：pip install paddlepaddle==2.0.0。
#### Q3.1.39: 字典中没有的字应该如何标注，是用空格代替还是直接忽略掉？
**A**：可以直接按照图片内容标注，在编码的时候，会忽略掉字典中不存在的字符。
#### Q3.1.40: dygraph、release/2.0-rc1-0、release/2.0 这三个分支有什么区别？
**A**：dygraph是动态图分支，并且适配Paddle-develop，当然目前在Paddle2.0上也可以运行，新特性我们会在这里更新。
release/2.0-rc1-0是基于Paddle 2.0rc1的稳定版本，release/2.0是基于Paddle2.0的稳定版本，如果希望版本或者代
码稳定的话，建议使用release/2.0分支，如果希望可以实时拿到一些最新特性，建议使用dygraph分支。
#### Q3.1.41: style-text 融合模块的输入是生成的前景图像以及背景特征权重吗？
**A**：目前版本是直接输入两个图像进行融合的，没有用到feature_map，替换背景图片不会影响效果。
#### Q3.1.42: 训练识别任务的时候，在CPU上运行时，报错`The setting of Parameter-Server must has server_num or servers`。
**A**：这是训练任务启动方式不对造成的。
1. 在使用CPU或者单块GPU训练的时候，可以直接使用`python3 tools/train.py -c xxx.yml`的方式启动。
2. 在使用多块GPU训练的时候，需要使用`distributed.launch`的方式启动，如`python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c xxx.yml`，这种方式需要安装NCCL库，如果没有的话会报错。
#### Q3.1.43：使用StyleText进行数据合成时，文本(TextInput)的长度远超StyleInput的长度，该怎么处理与合成呢？
**A**：在使用StyleText进行数据合成的时候，建议StyleInput的长度长于TextInput的长度。有2种方法可以处理上述问题：
1. 将StyleInput按列的方向进行复制与扩充，直到其超过TextInput的长度。
2. 将TextInput进行裁剪，保证每段TextInput都稍短于StyleInput，分别合成之后，再拼接在一起。
实际使用中发现，使用第2种方法的效果在长文本合成的场景中的合成效果更好，StyleText中提供的也是第2种数据合成的逻辑。
#### Q3.1.44: 文字识别训练，设置图像高度不等于32时报错
**A**：ctc decode的时候，输入需要是1维向量，因此降采样之后，建议特征图高度为1，ppocr中，特征图会降采样32倍，之后高度正好为1，所以有2种解决方案
- 指定输入shape高度为32（推荐）
- 在backbone的mv3中添加更多的降采样模块，保证输出的特征图高度为1
#### Q3.1.45: 增大batch_size模型训练速度没有明显提升
**A**：如果batch_size打得太大，加速效果不明显的话，可以试一下增大初始化内存的值，运行代码前设置环境变量：
```
export FLAGS_initial_cpu_memory_in_mb=2000 # 设置初始化内存约2G左右
```
#### Q3.1.46: 动态图分支(dygraph,release/2.0)，训练模型和推理模型效果不一致
**A**：当前问题表现为：使用训练完的模型直接测试结果较好，但是转换为inference model后，预测结果不一致；出现这个问题一般是两个原因：
1. 预处理函数设置的不一致
2. 后处理参数不一致
repo中config.yml文件的前后处理参数和inference预测默认的超参数有不一致的地方，建议排查下训练模型预测和inference预测的前后处理，
参考[issue](https://github.com/PaddlePaddle/PaddleOCR/issues/2080)。
#### Q3.1.47: paddleocr package 报错 FatalError: `Process abort signal` is detected by the operating system
**A**：首先，按照[安装文档](./installation.md)安装PaddleOCR的运行环境；另外，检查python环境，python3.6/3.8上可能会出现这个问题，建议用python3.7，
参考[issue](https://github.com/PaddlePaddle/PaddleOCR/issues/2069)。
#### Q3.1.48: 下载的识别模型解压后缺失文件，没有期望的inference.pdiparams, inference.pdmodel等文件
**A**：用解压软件解压可能会出现这个问题，建议二次解压下或者用命令行解压`tar xf `
#### Q3.1.49: 只想要识别票据中的部分片段，重新训练它的话，只需要训练文本检测模型就可以了吗？问文本识别，方向分类还是用原来的模型这样可以吗？
**A**：可以的。PaddleOCR的检测、识别、方向分类器三个模型是独立的，在实际使用中可以优化和替换其中任何一个模型。
#### Q3.1.50: 为什么在checkpoints中load下载的预训练模型会报错？
**A**: 这里有两个不同的概念：
- pretrained_model：指预训练模型，是已经训练完成的模型。这时会load预训练模型的参数，但并不会load学习率、优化器以及训练状态等。如果需要finetune，应该使用pretrained。
- checkpoints：指之前训练的中间结果，例如前一次训练到了100个epoch，想接着训练。这时会load尝试所有信息，包括模型的参数，之前的状态等。
这里应该使用pretrained_model而不是checkpoints
#### Q3.1.51: 如何用PaddleOCR识别视频中的文字？
**A**: 目前PaddleOCR主要针对图像做处理，如果需要视频识别，可以先对视频抽帧，然后用PPOCR识别。
#### Q3.1.52: 相机采集的图像为四通道，应该如何处理？
**A**: 有两种方式处理：
- 如果没有其他需要，可以在解码数据的时候指定模式为三通道，例如如果使用opencv，可以使用cv::imread(img_path, cv::IMREAD_COLOR)。
- 如果其他模块需要处理四通道的图像，那也可以在输入PaddleOCR模块之前进行转换，例如使用cvCvtColor(&img,img3chan,CV_RGBA2RGB)。
#### Q3.1.53: 预测时提示图像过大，显存、内存溢出了，应该如何处理？
**A**: 可以按照这个PR的修改来缓解显存、内存占用 [#2230](https://github.com/PaddlePaddle/PaddleOCR/pull/2230)
#### Q3.1.54: 用c++来部署，目前支持Paddle2.0的模型吗？
**A**: PPOCR 2.0的模型在arm上运行可以参照该PR [#1877](https://github.com/PaddlePaddle/PaddleOCR/pull/1877)
#### Q3.1.55: 目前PaddleOCR有知识蒸馏的demo吗？
**A**： 目前我们还没有提供PaddleOCR知识蒸馏的相关demo，PaddleClas开源了一个效果还不错的方案，可以移步[SSLD知识蒸馏方案](https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.0/docs/zh_CN/advanced_tutorials/distillation/distillation.md)， paper: https://arxiv.org/abs/2103.05959 关于PaddleOCR的蒸馏，我们也会在未来支持。
#### Q3.1.56: 在使用PPOCRLabel的时候，如何标注倾斜的文字？
**A**: 如果矩形框标注后空白冗余较多，可以尝试PPOCRLabel提供的四点标注，可以标注各种倾斜角度的文本。
#### Q3.1.57: 端到端算法PGNet提供了两种后处理方式，两者之间有什么区别呢？
**A**: 两种后处理的区别主要在于速度的推理，config中PostProcess有fast/slow两种模式，slow模式的后处理速度慢，精度相对较高，fast模式的后处理速度快，精度也在可接受的范围之内。建议使用速度快的后处理方式。
#### Q3.1.58: 使用PGNet进行eval报错？
**A**: 需要注意，我们目前在release/2.1更新了评测代码，目前支持A，B两种评测模式：
* A模式：该模式主要为了方便用户使用，与训练集一样的标注文件就可以正常进行eval操作, 代码中默认是A模式。
* B模式：该模式主要为了保证我们的评测代码可以和Total Text官方的评测方式对齐，该模式下直接加载官方提供的mat文件进行eval。
#### Q3.1.59: 使用预训练模型进行预测，对于特定字符识别识别效果较差，怎么解决？
**A**: 由于我们所提供的识别模型是基于通用大规模数据集进行训练的，部分字符可能在训练集中包含较少，因此您可以构建特定场景的数据集，基于我们提供的预训练模型进行微调。建议用于微调的数据集中，每个字符出现的样本数量不低于300，但同时需要注意不同字符的数量均衡。具体可以参考：[微调](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/recognition.md#2-%E5%90%AF%E5%8A%A8%E8%AE%AD%E7%BB%83)。
#### Q3.1.60: PGNet有中文预训练模型吗？
**A**: 目前我们尚未提供针对中文的预训练模型，如有需要，可以尝试自己训练。具体需要修改的地方有：
1. [config文件中](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/configs/e2e/e2e_r50_vd_pg.yml#L23-L24)，字典文件路径及语种设置；
1. [网络结构中](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/ppocr/modeling/heads/e2e_pg_head.py#L181)，`out_channels`修改为字典中的字符数目+1（考虑到空格）；
1. [loss中](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/ppocr/losses/e2e_pg_loss.py#L93)，修改`37`为字典中的字符数目+1（考虑到空格）；
#### Q3.1.61: 用于PGNet的训练集，文本框的标注有要求吗？
**A**: PGNet支持多点标注，比如4点、8点、14点等。但需要注意的是，标注点尽可能分布均匀（相邻标注点间隔距离均匀一致），且label文件中的标注点需要从标注框的左上角开始，按标注点顺时针顺序依次编写，以上问题都可能对训练精度造成影响。
我们提供的，基于Total Text数据集的PGNet预训练模型使用了14点标注方式。
#### Q3.1.62: 弯曲文本（如略微形变的文档图像）漏检问题
**A**: db后处理中计算文本框平均得分时，是求rectangle区域的平均分数，容易造成弯曲文本漏检，已新增求polygon区域的平均分数，会更准确，但速度有所降低，可按需选择，在相关pr中可查看[可视化对比效果](https://github.com/PaddlePaddle/PaddleOCR/pull/2604)。该功能通过参数 [det_db_score_mode](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/tools/infer/utility.py#L51)进行选择，参数值可选[`fast`(默认)、`slow`]，`fast`对应原始的rectangle方式，`slow`对应polygon方式。感谢用户[buptlihang](https://github.com/buptlihang)提[pr](https://github.com/PaddlePaddle/PaddleOCR/pull/2574)帮助解决该问题🌹。
#### Q3.1.63: 请问端到端的pgnet相比于DB+CRNN在准确率上有优势吗？或者是pgnet最擅长的场景是什么场景呢？
**A**: pgnet是端到端算法，检测识别一步到位，不用分开训练2个模型，也支持弯曲文本的识别，但是在中文上的效果还没有充分验证；db+crnn的验证更充分，应用相对成熟，常规非弯曲的文本都能解的不错。
#### Q3.1.64: config yml文件中的ratio_list参数的作用是什么？
**A**: 在动态图中，ratio_list在有多个数据源的情况下使用，ratio_list中的每个值是每个epoch从对应数据源采样数据的比例。如ratio_list=[0.3,0.2]，label_file_list=['data1','data2'],代表每个epoch的训练数据包含data1 30%的数据，和data2里 20%的数据，ratio_list中数值的和不需要等于1。ratio_list和label_file_list的长度必须一致。
静态图检测数据采样的逻辑与动态图不同，但基本不影响训练精度。
在静态图中，使用 检测 dataloader读取数据时，会先设置每个epoch的数据量，比如这里设置为1000，ratio_list中的值表示在1000中的占比，比如ratio_list是[0.3, 0.7]，则表示使用两个数据源，每个epoch从第一个数据源采样1000*0.3=300张图，从第二个数据源采样700张图。ratio_list的值的和也不需要等于1。
#### Q3.1.65: 支持动态图模型的android和ios demo什么时候上线？？
**A**: 支持动态图模型的android demo已经合入dygraph分支，欢迎试用（https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/deploy/android_demo/README.md）; ios demo暂时未提供动态图模型版本，可以基于静态图版本（https://github.com/PaddlePaddle/PaddleOCR/blob/develop/deploy/ios_demo）自行改造。
#### Q3.1.66: iaa里面添加的数据增强方式，是每张图像训练都会做增强还是随机的？如何添加一个数据增强方法？
**A**：iaa增强的训练配置参考：https://github.com/PaddlePaddle/PaddleOCR/blob/0ccc1720c252beb277b9e522a1b228eb6abffb8a/configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml#L82，
其中{ 'type': Fliplr, 'args': { 'p': 0.5 } } p是概率。新增数据增强，可以参考这个方法：https://github.com/PaddlePaddle/PaddleOCR/blob/release%2F2.1/doc/doc_ch/add_new_algorithm.md#%E6%95%B0%E6%8D%AE%E5%8A%A0%E8%BD%BD%E5%92%8C%E5%A4%84%E7%90%86
#### Q3.1.67: PGNet训练中文弯曲数据集，可视化时弯曲文本无法显示。
**A**: 可能是因为安装的OpenCV里，cv2.putText不能显示中文的原因，可以尝试用Pillow来添加显示中文，需要改draw_e2e_res函数里面的代码，可以参考如下代码：
```
box = box.astype(np.int32).reshape((-1, 1, 2))
cv2.polylines(src_im, [box], True, color=(255, 255, 0), thickness=2)
from PIL import ImageFont, ImageDraw, Image
img = Image.fromarray(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
draw = ImageDraw.Draw(img)
fontStyle = ImageFont.truetype(
"font/msyh.ttc", 16, encoding="utf-8")
draw.text((int(box[0, 0, 0]), int(box[0, 0, 1])), text, (0, 255, 0), font=fontStyle)
src_im= cv2.cvtColor(np.asarray(img), cv2.COLOR_RGB2BGR)
```
#### Q3.1.68: 用PGNet做进行端到端训练时，数据集标注的点的个数必须都是统一一样的吗? 能不能随意标点数，只要能够按顺时针从左上角开始标这样?
**A**: 目前代码要求标注为统一的点数。
#### Q3.1.69: 怎么加速训练过程呢？
**A**：OCR模型训练过程中一般包含大量的数据增广，这些数据增广是比较耗时的，因此可以离线生成大量增广后的图像，直接送入网络进行训练，机器资源充足的情况下，也可以使用分布式训练的方法，可以参考[分布式训练教程文档](https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/doc/doc_ch/distributed_training.md)。
#### Q3.1.70: 文字识别模型模型的输出矩阵需要进行解码才能得到识别的文本。代码中实现为preds_idx = preds.argmax(axis=2)，也就是最佳路径解码法。这是一种贪心算法，是每一个时间步只将最大概率的字符作为当前时间步的预测输出，但得到的结果不一定是最好的。为什么不使用beam search这种方式进行解码呢？
**A**：实验发现，使用贪心的方法去做解码，识别精度影响不大，但是速度方面的优势比较明显，因此PaddleOCR中使用贪心算法去做识别的解码。
#### Q3.1.71: 遇到中英文识别模型不支持的字符，该如何对模型做微调？
**A**：如果希望识别中英文识别模型中不支持的字符，需要更新识别的字典，并完成微调过程。比如说如果希望模型能够进一步识别罗马数字，可以按照以下步骤完成模型微调过程。
1. 准备中英文识别数据以及罗马数字的识别数据，用于训练，同时保证罗马数字和中英文识别数字的效果；
2. 修改默认的字典文件，在后面添加罗马数字的字符；
3. 下载PaddleOCR提供的预训练模型，配置预训练模型和数据的路径，开始训练。
#### Q3.1.72: 文字识别主要有CRNN和Attention两种方式，但是在我们的说明文档中，CRNN有对应的论文，但是Attention没看到，这个具体在哪里呢？
**A**：文字识别主要有CTC和Attention两种方式，基于CTC的算法有CRNN、Rosetta、StarNet，基于Attention的方法有RARE、其他的算法PaddleOCR里没有提供复现代码。论文的链接可以参考：[PaddleOCR文本识别算法教程文档](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/algorithm_overview.md#%E6%96%87%E6%9C%AC%E8%AF%86%E5%88%AB%E7%AE%97%E6%B3%95)
#### Q3.1.73: 如何使用TensorRT加速PaddleOCR预测？
**A**： 目前paddle的dygraph分支已经支持了python和C++ TensorRT预测的代码，python端inference预测时把参数[--use_tensorrt=True](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/tools/infer/utility.py#L37)即可，
C++TensorRT预测需要使用支持TRT的预测库并在编译时打开[-DWITH_TENSORRT=ON](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/deploy/cpp_infer/tools/build.sh#L15)。
如果想修改其他分支代码支持TensorRT预测，可以参考[PR](https://github.com/PaddlePaddle/PaddleOCR/pull/2921)。
注：建议使用TensorRT大于等于6.1.0.5以上的版本。
#### Q3.1.74: ppocr检测效果不好，该如何优化？
**A**： 具体问题具体分析:
1. 如果在你的场景上检测效果不可用，首选是在你的数据上做finetune训练；
2. 如果图像过大，文字过于密集，建议不要过度压缩图像，可以尝试修改检测预处理的[resize逻辑](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/tools/infer/predict_det.py#L42)，防止图像被过度压缩；
3. 检测框大小过于紧贴文字或检测框过大，可以调整[db_unclip_ratio](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/tools/infer/utility.py#L51)这个参数，加大参数可以扩大检测框，减小参数可以减小检测框大小；
4. 检测框存在很多漏检问题，可以减小DB检测后处理的阈值参数[det_db_box_thresh](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/tools/infer/utility.py#L50)，防止一些检测框被过滤掉，也可以尝试设置[det_db_score_mode](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/tools/infer/utility.py#L54)为'slow';
5. 其他方法可以选择[use_dilation](https://github.com/PaddlePaddle/PaddleOCR/blob/3ec57e8df9263de6fa897e33d2d91bc5d0849ef3/tools/infer/utility.py#L53)为True，对检测输出的feature map做膨胀处理，一般情况下，会有效果改善；
#### Q3.1.75: lite预测库和nb模型版本不匹配，该如何解决？
**A**： 如果可以正常预测就不用管，如果这个问题导致无法正常预测，可以尝试使用同一个commit的Paddle Lite代码编译预测库和opt文件，可以参考[移动端部署教程](https://github.com/PaddlePaddle/PaddleOCR/blob/release%2F2.1/deploy/lite/readme.md)。
#### Q3.1.76: 'SystemError: (Fatal) Blocking queue is killed because the data reader raises an exception.' 遇到这个错如何处理？
这个报错说明dataloader的时候报错了，如果是还未开始训练就报错，需要检查下数据和标签格式是不是对的，ppocr的数据标签格式为
```
" 图像文件名 json.dumps编码的图像标注信息"
ch4_test_images/img_61.jpg [{"transcription": "MASA", "points": [[310, 104], [416, 141], [418, 216], [312, 179]]}, {...}]
```
提供的标注文件格式如上，中间用"\t"分隔，不是四个空格分隔。
如果是训练期间报错了，需要检查下是不是遇到了异常数据，或者是共享内存不足导致了这个问题，可以使用tools/train.py中的test_reader进行调试，
linux系统共享内存位于/dev/shm目录下，如果内存不足，可以清理/dev/shm目录, 另外，如果是使用docker，在创建镜像时，可通过设置参数--shm_size=8G 设置较大的共享内存。
#### Q3.1.77: 使用mkldnn加速预测时遇到 'Please compile with MKLDNN first to use MKLDNN'
**A**： 报错提示当前环境没有mkldnn，建议检查下当前CPU是否支持mlkdnn（MAC上是无法用mkldnn）；另外的可能是使用的预测库不支持mkldnn，
建议从[这里](https://paddle-inference.readthedocs.io/en/latest/user_guides/download_lib.html#linux)下载支持mlkdnn的CPU预测库。
#### Q3.1.78: 在线demo支持阿拉伯语吗
**A**： 在线demo目前只支持中英文， 多语言的都需要通过whl包自行处理
#### Q3.1.79: 某个类别的样本比较少，通过增加训练的迭代次数或者是epoch，变相增加小样本的数目，这样能缓解这个问题么？
**A**： 尽量保证类别均衡， 某些类别样本少，可以通过补充合成数据的方式处理；实验证明训练集中出现频次较少的字符，识别效果会比较差，增加迭代次数不能改变样本量少的问题。
#### Q3.1.80: 想把简历上的文字识别出来后，能够把关系一一对应起来，比如姓名和它后面的名字组成一对，籍贯、邮箱、学历等等都和各自的内容关联起来，这个应该如何处理，PPOCR目前支持吗？
**A**: 这样的需求在企业应用中确实比较常见，但往往都是个性化的需求，没有非常规整统一的处理方式。常见的处理方式有如下两种：
1. 对于单一版式、或者版式差异不大的应用场景，可以基于识别场景的一些先验信息，将识别内容进行配对； 比如运用表单结构信息：常见表单"姓名"关键字的后面，往往紧跟的就是名字信息
2. 对于版式多样，或者无固定版式的场景， 需要借助于NLP中的NER技术，给识别内容中的某些字段，赋予key值
由于这部分需求和业务场景强相关，难以用一个统一的模型去处理，目前PPOCR暂不支持。 如果需要用到NER技术，可以参照Paddle团队的另一个开源套件: https://github.com/PaddlePaddle/ERNIE， 其提供的预训练模型ERNIE, 可以帮助提升NER任务的准确率。
<a name="数据集3"></a>
### 数据集
#### Q3.2.1：如何制作PaddleOCR支持的数据格式
......@@ -576,6 +881,10 @@ StyleText的用途主要是：提取style_image中的字体、背景等style信
#### Q3.2.18: PaddleOCR动态图版本如何finetune？
**A**：finetune需要将配置文件里的 Global.load_static_weights设置为false，如果没有此字段可以手动添加，然后将模型地址放到Global.pretrained_model字段下即可。
#### Q3.2.19: 如何合成手写中文数据集？
**A**: 手写数据集可以通过手写单字数据集合成得到。随机选取一定数量的单字图片和对应的label，将图片高度resize为随机的统一高度后拼接在一起，即可得到合成数据集。对于需要添加文字背景的情况，建议使用阈值化将单字图片的白色背景处理为透明背景，再与真实背景图进行合成。具体可以参考文档[手写数据集](https://github.com/PaddlePaddle/PaddleOCR/blob/a72d6f23be9979e0c103d911a9dca3e4613e8ccf/doc/doc_ch/handwritten_datasets.md)。
<a name="模型训练调优3"></a>
### 模型训练调优
......@@ -725,8 +1034,52 @@ ps -axu | grep train.py | awk '{print $2}' | xargs kill -9
**A**：1.1和2.0的模型一样，微调时，垂直排列的文字需要逆时针旋转 90°后加入训练，上下颠倒的需要旋转为水平的。
#### Q3.3.30: 模型训练过程中如何得到 best_accuracy 模型？
**A**：配置文件里的eval_batch_step字段用来控制多少次iter进行一次eval，在eval完成后会自动生成 best_accuracy 模型，所以如果希望很快就能拿到best_accuracy模型，可以将eval_batch_step改小一点，如改为[10,10]，这样表示第10次迭代后，以后没隔10个迭代就进行一次模型的评估。
#### Q3.3.31: Cosine学习率的更新策略是怎样的？训练过程中为什么会在一个值上停很久？
**A**: Cosine学习率的说明可以参考[这里](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/optimizer/lr/CosineAnnealingDecay_cn.html#cosineannealingdecay)
在PaddleOCR中，为了让学习率更加平缓，我们将其中的epoch调整成了iter。
学习率的更新会和总的iter数量有关。当iter比较大时，会经过较多iter才能看出学习率的值有变化。
#### Q3.3.32: 之前的CosineWarmup方法为什么不见了？
**A**: 我们对代码结构进行了调整，目前的Cosine可以覆盖原有的CosineWarmup的功能，只需要在配置文件中增加相应配置即可。
例如下面的代码，可以设置warmup为2个epoch：
```
lr:
name: Cosine
learning_rate: 0.001
warmup_epoch: 2
```
#### Q3.3.33: 训练识别和检测时学习率要加上warmup，目的是什么？
**A**: Warmup机制先使学习率从一个较小的值逐步升到一个较大的值，而不是直接就使用较大的学习率，这样有助于模型的稳定收敛。在OCR检测和OCR识别中，一般会带来精度~0.5%的提升。
#### Q3.3.34: 表格识别中，如何提高单字的识别结果？
**A**: 首先需要确认一下检测模型有没有有效的检测出单个字符，如果没有的话，需要在训练集当中添加相应的单字数据集。
#### Q3.3.35: SRN训练不收敛（loss不降）或SRN训练acc一直为0。
**A**: 如果loss下降不正常，需要确认没有修改yml文件中的image_shape，默认[1, 64, 256]，代码中针对这个配置写死了，修改可能会造成无法收敛。如果确认参数无误，loss正常下降，可以多迭代一段时间观察下，开始acc为0是正常的。
#### Q3.3.36: 训练starnet网络，印章数据可以和非弯曲数据一起训练吗。
**A**: 可以的，starnet里的tps模块会对印章图片进行校正，使其和非弯曲的图片一样。
#### Q3.3.37: 训练过程中，训练程序意外退出/挂起，应该如何解决？
**A**: 考虑内存，显存（使用GPU训练的话）是否不足，可在配置文件中，将训练和评估的batch size调小一些。需要注意，训练batch size调小时，学习率learning rate也要调小，一般可按等比例调整。
#### Q3.3.38: 训练程序启动后直到结束，看不到训练过程log？
**A**: 可以从以下三方面考虑：
1. 检查训练进程是否正常退出、显存占用是否释放、是否有残留进程，如果确定是训练程序卡死，可以检查环境配置，遇到环境问题建议使用docker，可以参考说明文档[安装](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/installation.md)。
2. 检查数据集的数据量是否太小，可调小batch size从而增加一个epoch中的训练step数量，或在训练config文件中，将参数print_batch_step改为1，即每一个step打印一次log信息。
3. 如果使用私有数据集训练，可先用PaddleOCR提供/推荐的数据集进行训练，排查私有数据集是否存在问题。
#### Q3.3.39: 配置文件中的参数num workers是什么意思，应该如何设置？
**A**: 训练数据的读取需要硬盘IO，而硬盘IO速度远小于GPU运算速度，为了避免数据读取成为训练速度瓶颈，可以使用多进程读取数据，num workers表示数据读取的进程数量，0表示不使用多进程读取。在Linux系统下，多进程读取数据时，进程间通信需要基于共享内存，因此使用多进程读取数据时，建议设置共享内存不低于2GB，最好可以达到8GB，此时，num workers可以设置为CPU核心数。如果机器硬件配置较低，或训练进程卡死、dataloader报错，可以将num workers设置为0，即不使用多进程读取数据。
<a name="预测部署3"></a>
### 预测部署
......@@ -771,10 +1124,6 @@ ps -axu | grep train.py | awk '{print $2}' | xargs kill -9
**A**：在安卓APK上无法设置，没有暴露这个接口，如果使用的是PaddledOCR/deploy/lite/的demo，可以修改config.txt中的对应参数来设置
#### Q3.4.9：PaddleOCR模型是否可以转换成ONNX模型?
**A**：目前暂不支持转ONNX，相关工作在研发中。
#### Q3.4.10：使用opt工具对检测模型转换时报错 can not found op arguments for node conv2_b_attr
**A**：这个问题大概率是编译opt工具的Paddle-Lite不是develop分支，建议使用Paddle-Lite 的develop分支编译opt工具。
......@@ -841,7 +1190,8 @@ ps -axu | grep train.py | awk '{print $2}' | xargs kill -9
**A**：使用EAST或SAST模型进行推理预测时，需要在命令中指定参数--det_algorithm="EAST" 或 --det_algorithm="SAST"，使用DB时不用指定是因为该参数默认值是"DB"：https://github.com/PaddlePaddle/PaddleOCR/blob/e7a708e9fdaf413ed7a14da8e4a7b4ac0b211e42/tools/infer/utility.py#L43
#### Q3.4.25: PaddleOCR模型Python端预测和C++预测结果不一致？
正常来说，python端预测和C++预测文本是一致的，如果预测结果差异较大，
**A**：正常来说，python端预测和C++预测文本是一致的，如果预测结果差异较大，
建议首先排查diff出现在检测模型还是识别模型，或者尝试换其他模型是否有类似的问题。
其次，检查python端和C++端数据处理部分是否存在差异，建议保存环境，更新PaddleOCR代码再试下。
如果更新代码或者更新代码都没能解决，建议在PaddleOCR微信群里或者issue中抛出您的问题。
......@@ -889,3 +1239,68 @@ Paddle2ONNX支持转换的[模型列表](https://github.com/PaddlePaddle/Paddle2
#### Q3.4.34: 2.0训练出来的模型，能否在1.1版本上进行部署？
**A**：这个是不建议的，2.0训练出来的模型建议使用dygraph分支里提供的部署代码。
#### Q3.4.35: 怎么解决paddleOCR在T4卡上有越预测越慢的情况？
**A**：
1. T4 GPU没有主动散热，因此在测试的时候需要在每次infer之后需要sleep 30ms，否则机器容易因为过热而降频(inference速度会变慢)，温度过高也有可能会导致宕机。
2. T4在不使用的时候，也有可能会降频，因此在做benchmark的时候需要锁频，下面这两条命令可以进行锁频。
```
nvidia-smi -i 0 -pm ENABLED
nvidia-smi --lock-gpu-clocks=1590 -i 0
```
#### Q3.4.36: DB有些框太贴文本了反而去掉了一些文本的边角影响识别，这个问题有什么办法可以缓解吗？
**A**：可以把后处理的参数unclip_ratio适当调大一点。
#### Q3.4.37: 在windows上进行cpp inference的部署时，总是提示找不到`paddle_fluid.dll`和`opencv_world346.dll`，
**A**：有2种方法可以解决这个问题：
1. 将paddle预测库和opencv库的地址添加到系统环境变量中。
2. 将提示缺失的dll文件拷贝到编译产出的`ocr_system.exe`文件夹中。
#### Q3.4.38：想在Mac上部署，从哪里下载预测库呢？
**A**：Mac上的Paddle预测库可以从这里下载：[https://paddle-inference-lib.bj.bcebos.com/mac/2.0.0/cpu_avx_openblas/paddle_inference.tgz](https://paddle-inference-lib.bj.bcebos.com/mac/2.0.0/cpu_avx_openblas/paddle_inference.tgz)
#### Q3.4.39：内网环境如何进行服务化部署呢？
**A**：仍然可以使用PaddleServing或者HubServing进行服务化部署，保证内网地址可以访问即可。
#### Q3.4.40: 使用hub_serving部署，延时较高，可能的原因是什么呀？
**A**: 首先，测试的时候第一张图延时较高，可以多测试几张然后观察后几张图的速度；其次，如果是在cpu端部署serving端模型（如backbone为ResNet34），耗时较慢，建议在cpu端部署mobile（如backbone为MobileNetV3）模型。
#### Q3.4.41: PaddleOCR支持tensorrt推理吗？
**A**: 支持的，需要在编译的时候将CMakeLists.txt文件当中，将相关代码`option(WITH_TENSORRT "Compile demo with TensorRT." OFF)`的OFF改成ON。关于服务器端部署的更多设置，可以参考[飞桨官网](https://www.paddlepaddle.org.cn/documentation/docs/zh/guides/05_inference_deployment/inference/native_infer.html)
#### Q3.4.42: 在使用PaddleLite进行预测部署时，启动预测后卡死/手机死机？
**A**: 请检查模型转换时所用PaddleLite的版本，和预测库的版本是否对齐。即PaddleLite版本为2.8，则预测库版本也要为2.8。
#### Q3.4.43: 预测时显存爆炸、内存泄漏问题？
**A**: 打开显存/内存优化开关`enable_memory_optim`可以解决该问题，相关代码已合入，[查看详情](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/tools/infer/utility.py#L153)。
#### Q3.4.44: 如何多进程预测？
**A**: 近期PaddleOCR新增了[多进程预测控制参数](https://github.com/PaddlePaddle/PaddleOCR/blob/a312647be716776c1aac33ff939ae358a39e8188/tools/infer/utility.py#L103)，`use_mp`表示是否使用多进程，`total_process_num`表示在使用多进程时的进程数。具体使用方式请参考[文档](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/inference.md#1-%E8%B6%85%E8%BD%BB%E9%87%8F%E4%B8%AD%E6%96%87ocr%E6%A8%A1%E5%9E%8B%E6%8E%A8%E7%90%86)。
#### Q3.4.45: win下C++部署中文识别乱码的解决方法
**A**: win下编码格式不是utf8,而ppocr_keys_v1.txt的编码格式的utf8，将ppocr_keys_v1.txt 的编码从utf-8修改为 Ansi 编码格式就行了。
#### Q3.4.46: windows 3060显卡GPU模式启动 加载模型慢。
**A**: 30系列的显卡需要使用cuda11。
#### Q3.4.47: 请教如何优化检测阶段时长?
**A**: 预测单张图会慢一点，如果批量预测，第一张图比较慢，后面就快了，因为最开始一些初始化操作比较耗时。服务部署的话，访问一次后，后面再访问就不会初始化了，推理的话每次都需要初始化的。
#### Q3.4.48: paddle serving 本地启动调用失败,怎么判断是否正常工作？
**A**：没有打印出预测结果，说明启动失败。可以参考这篇文档重新配置下动态图的paddle serving：https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/deploy/pdserving/README_CN.md
#### Q3.4.49: 同一个模型，c++部署和python部署方式，出来的结果不一致，如何定位？
**A**：有如下几个Debug经验：
1. 优先对一下几个阈值参数是否一致；
2. 排查一下c++代码和python代码的预处理和后处理方式是否一致；
3. 用python在模型输入输出各保存一下二进制文件，排除inference的差异性
doc/doc_ch/config.md
浏览文件 @ f57a6d1b
## 可选参数列表
# 配置文件内容与生成
## 1. 可选参数列表
以下列表可以通过`--help`查看
......@@ -8,10 +10,10 @@
| -o | ALL | 设置配置文件里的参数内容 | None | 使用-o配置相较于-c选择的配置文件具有更高的优先级。例如：`-o Global.use_gpu=false` |
## 配置文件参数介绍
## 2. 配置文件参数介绍
以 `rec_chinese_lite_train_v2.0.yml ` 为例
### Global
### 2.1 Global
| 字段 | 用途 | 默认值 | 备注 |
| :----------------------: | :---------------------: | :--------------: | :--------------------: |
......@@ -52,7 +54,7 @@
### Architecture ([ppocr/modeling](../../ppocr/modeling))
在ppocr中，网络被划分为Transform,Backbone,Neck和Head四个阶段
在PaddleOCR中，网络被划分为Transform,Backbone,Neck和Head四个阶段
| 字段 | 用途 | 默认值 | 备注 |
| :---------------------: | :---------------------: | :--------------: | :--------------------: |
......@@ -121,3 +123,97 @@
| batch_size_per_card | 训练时单卡batch size | 256 | \ |
| drop_last | 是否丢弃因数据集样本数不能被 batch_size 整除而产生的最后一个不完整的mini-batch | True | \ |
| num_workers | 用于加载数据的子进程个数，若为0即为不开启子进程，在主进程中进行数据加载 | 8 | \ |
## 3. 多语言配置文件生成
PaddleOCR目前已支持80种（除中文外）语种识别，`configs/rec/multi_languages` 路径下提供了一个多语言的配置文件模版: [rec_multi_language_lite_train.yml](../../configs/rec/multi_language/rec_multi_language_lite_train.yml)。
您有两种方式创建所需的配置文件：
1. 通过脚本自动生成
[generate_multi_language_configs.py](../../configs/rec/multi_language/generate_multi_language_configs.py) 可以帮助您生成多语言模型的配置文件
- 以意大利语为例，如果您的数据是按如下格式准备的：
```
|-train_data
|- it_train.txt # 训练集标签
|- it_val.txt # 验证集标签
|- data
|- word_001.jpg
|- word_002.jpg
|- word_003.jpg
| ...
```
可以使用默认参数，生成配置文件：
```bash
# 该代码需要在指定目录运行
cd PaddleOCR/configs/rec/multi_language/
# 通过-l或者--language参数设置需要生成的语种的配置文件，该命令会将默认参数写入配置文件
python3 generate_multi_language_configs.py -l it
```
- 如果您的数据放置在其他位置，或希望使用自己的字典，可以通过指定相关参数来生成配置文件:
```bash
# -l或者--language字段是必须的
# --train修改训练集，--val修改验证集，--data_dir修改数据集目录，--dict修改字典路径， -o修改对应默认参数
cd PaddleOCR/configs/rec/multi_language/
python3 generate_multi_language_configs.py -l it \ # 语种
--train {path/of/train_label.txt} \ # 训练标签文件的路径
--val {path/of/val_label.txt} \ # 验证集标签文件的路径
--data_dir {train_data/path} \ # 训练数据的根目录
--dict {path/of/dict} \ # 字典文件路径
-o Global.use_gpu=False # 是否使用gpu
...
```
意大利文由拉丁字母组成，因此执行完命令后会得到名为 rec_latin_lite_train.yml 的配置文件。
2. 手动修改配置文件
您也可以手动修改模版中的以下几个字段得到配置文件:
```
Global:
use_gpu: True
epoch_num: 500
...
character_type: it # 需要识别的语种
character_dict_path: {path/of/dict} # 字典文件所在路径
Train:
dataset:
name: SimpleDataSet
data_dir: train_data/ # 数据存放根目录
label_file_list: ["./train_data/train_list.txt"] # 训练集label路径
...
Eval:
dataset:
name: SimpleDataSet
data_dir: train_data/ # 数据存放根目录
label_file_list: ["./train_data/val_list.txt"] # 验证集label路径
...
```
目前PaddleOCR支持的多语言算法有：
| 配置文件 | 算法名称 | backbone | trans | seq | pred | language | character_type |
| :--------: | :-------: | :-------: | :-------: | :-----: | :-----: | :-----: | :-----: |
| rec_chinese_cht_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 中文繁体 | chinese_cht|
| rec_en_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 英语（区分大小写） | EN |
| rec_french_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 法语 | french |
| rec_ger_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 德语 | german |
| rec_japan_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 日语 | japan |
| rec_korean_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 韩语 | korean |
| rec_latin_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 拉丁字母 | latin |
| rec_arabic_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 阿拉伯字母 | ar |
| rec_cyrillic_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 斯拉夫字母 | cyrillic |
| rec_devanagari_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | 梵文字母 | devanagari |
更多支持语种请参考: [多语言模型](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/multi_languages.md#%E8%AF%AD%E7%A7%8D%E7%BC%A9%E5%86%99)
doc/doc_ch/detection.md
浏览文件 @ f57a6d1b
# 文字检测
本节以icdar2015数据集为例，介绍PaddleOCR中检测模型的训练、评估与测试。
# 目录
- [1. 文字检测](#1-----)
* [1.1 数据准备](#11-----)
* [1.2 下载预训练模型](#12--------)
* [1.3 启动训练](#13-----)
* [1.4 断点训练](#14-----)
* [1.5 更换Backbone 训练](#15---backbone---)
* [1.6 指标评估](#16-----)
* [1.7 测试检测效果](#17-------)
* [1.8 转inference模型测试](#18--inference----)
- [2. FAQ](#2-faq)
## 数据准备
<a name="1-----"></a>
# 1. 文字检测
本节以icdar2015数据集为例，介绍PaddleOCR中检测模型训练、评估、测试的使用方式。
<a name="11-----"></a>
## 1.1 数据准备
icdar2015数据集可以从[官网](https://rrc.cvc.uab.es/?ch=4&com=downloads)下载到，首次下载需注册。
注册完成登陆后，下载下图中红色框标出的部分，其中， `Training Set Images`下载的内容保存为`icdar_c4_train_imgs`文件夹下，`Test Set Images` 下载的内容保存为`ch4_test_images`文件夹下
<p align="center">
<img src="./doc/datasets/ic15_location_download.png" align="middle" width = "600"/>
<p align="center">
将下载到的数据集解压到工作目录下，假设解压在 PaddleOCR/train_data/ 下。另外，PaddleOCR将零散的标注文件整理成单独的标注文件
，您可以通过wget的方式进行下载。
```shell
......@@ -23,7 +45,7 @@ python gen_label.py --mode="det" --root_path="/path/to/icdar_c4_train_imgs/" \
--output_label="/path/to/train_icdar2015_label.txt"
```
解压数据集和下载标注文件后，PaddleOCR/train_data/ 有两个文件夹和两个文件，分别是：
解压数据集和下载标注文件后，PaddleOCR/train_data/ 有两个文件夹和两个文件，按照如下方式组织icdar2015数据集：
```
/PaddleOCR/train_data/icdar2015/text_localization/
└─ icdar_c4_train_imgs/ icdar数据集的训练数据
......@@ -42,11 +64,13 @@ json.dumps编码前的图像标注信息是包含多个字典的list，字典中
如果您想在其他数据集上训练，可以按照上述形式构建标注文件。
## 快速启动训练
<a name="12--------"></a>
## 1.2 下载预训练模型
首先下载模型backbone的pretrain model，PaddleOCR的检测模型目前支持两种backbone，分别是MobileNetV3、ResNet_vd系列，
您可以根据需求使用[PaddleClas](https://github.com/PaddlePaddle/PaddleClas/tree/develop/ppcls/modeling/architectures)中的模型更换backbone，
对应的backbone预训练模型可以从[PaddleClas repo 主页中找到下载链接](https://github.com/PaddlePaddle/PaddleClas#mobile-series)。
您可以根据需求使用[PaddleClas](https://github.com/PaddlePaddle/PaddleClas/tree/release/2.0/ppcls/modeling/architectures)中的模型更换backbone，
对应的backbone预训练模型可以从[PaddleClas repo 主页中找到下载链接](https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.0/README_cn.md#resnet%E5%8F%8A%E5%85%B6vd%E7%B3%BB%E5%88%97)。
```shell
cd PaddleOCR/
# 根据backbone的不同选择下载对应的预训练模型
......@@ -56,23 +80,23 @@ wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dyg
wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet18_vd_pretrained.pdparams
# 或，下载ResNet50_vd的预训练模型
wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams
```
#### 启动训练
<a name="13-----"></a>
## 1.3 启动训练
*如果您安装的是cpu版本，请将配置文件中的 `use_gpu` 字段修改为false*
```shell
# 单机单卡训练 mv3_db 模型
python3 tools/train.py -c configs/det/det_mv3_db.yml \
-o Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained/
-o Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained
# 单机多卡训练，通过 --gpus 参数设置使用的GPU ID
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/det/det_mv3_db.yml \
-o Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained/
-o Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained
```
上述指令中，通过-c 选择训练使用configs/det/det_db_mv3.yml配置文件。
有关配置文件的详细解释，请参考[链接](./config.md)。
......@@ -81,46 +105,122 @@ python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/
python3 tools/train.py -c configs/det/det_mv3_db.yml -o Optimizer.base_lr=0.0001
```
#### 断点训练
<a name="14-----"></a>
## 1.4 断点训练
如果训练程序中断，如果希望加载训练中断的模型从而恢复训练，可以通过指定Global.checkpoints指定要加载的模型路径：
```shell
python3 tools/train.py -c configs/det/det_mv3_db.yml -o Global.checkpoints=./your/trained/model
```
**注意**：`Global.checkpoints`的优先级高于`Global.pretrain_weights`的优先级，即同时指定两个参数时，优先加载`Global.checkpoints`指定的模型，如果`Global.checkpoints`指定的模型路径有误，会加载`Global.pretrain_weights`指定的模型。
<a name="15---backbone---"></a>
## 1.5 更换Backbone 训练
PaddleOCR将网络划分为四部分，分别在[ppocr/modeling](../../ppocr/modeling)下。 进入网络的数据将按照顺序(transforms->backbones->
necks->heads)依次通过这四个部分。
```bash
├── architectures # 网络的组网代码
├── transforms # 网络的图像变换模块
├── backbones # 网络的特征提取模块
├── necks # 网络的特征增强模块
└── heads # 网络的输出模块
```
如果要更换的Backbone 在PaddleOCR中有对应实现，直接修改配置yml文件中`Backbone`部分的参数即可。
**注意**：`Global.checkpoints`的优先级高于`Global.pretrain_weights`的优先级，即同时指定两个参数时，优先加载`Global.checkpoints`指定的模型，如果`Global.checkpoints`指定的模型路径有误，会加载`Global.pretrain_weights`指定的模型。
如果要使用新的Backbone，更换backbones的例子如下:
1. 在 [ppocr/modeling/backbones](../../ppocr/modeling/backbones) 文件夹下新建文件，如my_backbone.py。
2. 在 my_backbone.py 文件内添加相关代码，示例代码如下:
```python
import paddle
import paddle.nn as nn
import paddle.nn.functional as F
class MyBackbone(nn.Layer):
def __init__(self, *args, **kwargs):
super(MyBackbone, self).__init__()
# your init code
self.conv = nn.xxxx
def forward(self, inputs):
# your network forward
y = self.conv(inputs)
return y
```
3. 在 [ppocr/modeling/backbones/\__init\__.py](../../ppocr/modeling/backbones/__init__.py)文件内导入添加的`MyBackbone`模块，然后修改配置文件中Backbone进行配置即可使用，格式如下:
## 指标评估
```yaml
Backbone:
name: MyBackbone
args1: args1
```
PaddleOCR计算三个OCR检测相关的指标，分别是：Precision、Recall、Hmean。
**注意**：如果要更换网络的其他模块，可以参考[文档](./add_new_algorithm.md)。
运行如下代码，根据配置文件`det_db_mv3.yml`中`save_res_path`指定的测试集检测结果文件，计算评估指标。
<a name="16-----"></a>
## 1.6 指标评估
PaddleOCR计算三个OCR检测相关的指标，分别是：Precision、Recall、Hmean（F-Score）。
评估时设置后处理参数`box_thresh=0.5`，`unclip_ratio=1.5`，使用不同数据集、不同模型训练，可调整这两个参数进行优化
训练中模型参数默认保存在`Global.save_model_dir`目录下。在评估指标时，需要设置`Global.checkpoints`指向保存的参数文件。
```shell
python3 tools/eval.py -c configs/det/det_mv3_db.yml -o Global.checkpoints="{path/to/weights}/best_accuracy" PostProcess.box_thresh=0.5 PostProcess.unclip_ratio=1.5
python3 tools/eval.py -c configs/det/det_mv3_db.yml -o Global.checkpoints="{path/to/weights}/best_accuracy"
```
* 注：`box_thresh`、`unclip_ratio`是DB后处理所需要的参数，在评估EAST模型时不需要设置
## 测试检测效果
<a name="17-------"></a>
## 1.7 测试检测效果
测试单张图像的检测效果
```shell
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/img_10.jpg" Global.pretrained_model="./output/det_db/best_accuracy"
```
测试DB模型时，调整后处理阈值，
测试DB模型时，调整后处理阈值
```shell
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/img_10.jpg" Global.pretrained_model="./output/det_db/best_accuracy" PostProcess.box_thresh=0.6 PostProcess.unclip_ratio=1.5
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/img_10.jpg" Global.pretrained_model="./output/det_db/best_accuracy" PostProcess.box_thresh=0.6 PostProcess.unclip_ratio=2.0
```
测试文件夹下所有图像的检测效果
```shell
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/" Global.pretrained_model="./output/det_db/best_accuracy"
```
<a name="18--inference----"></a>
## 1.8 转inference模型测试
inference 模型（`paddle.jit.save`保存的模型）
一般是模型训练，把模型结构和模型参数保存在文件中的固化模型，多用于预测部署场景。
训练过程中保存的模型是checkpoints模型，保存的只有模型的参数，多用于恢复训练等。
与checkpoints模型相比，inference 模型会额外保存模型的结构信息，在预测部署、加速推理上性能优越，灵活方便，适合于实际系统集成。
检测模型转inference 模型方式：
```shell
# 加载配置文件`det_mv3_db.yml`，从`output/det_db`目录下加载`best_accuracy`模型，inference模型保存在`./output/det_db_inference`目录下
python3 tools/export_model.py -c configs/det/det_mv3_db.yml -o Global.pretrained_model="./output/det_db/best_accuracy" Global.save_inference_dir="./output/det_db_inference/"
```
DB检测模型inference 模型预测：
```shell
python3 tools/infer/predict_det.py --det_algorithm="DB" --det_model_dir="./output/det_db_inference/" --image_dir="./doc/imgs/" --use_gpu=True
```
如果是其他检测，比如EAST模型，det_algorithm参数需要修改为EAST，默认为DB算法：
```shell
python3 tools/infer/predict_det.py --det_algorithm="EAST" --det_model_dir="./output/det_db_inference/" --image_dir="./doc/imgs/" --use_gpu=True
```
<a name="2"></a>
# 2. FAQ
Q1: 训练模型转inference 模型之后预测效果不一致？
**A**：此类问题出现较多，问题多是trained model预测时候的预处理、后处理参数和inference model预测的时候的预处理、后处理参数不一致导致的。以det_mv3_db.yml配置文件训练的模型为例，训练模型、inference模型预测结果不一致问题解决方式如下：
- 检查[trained model预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L116)，和[inference model的预测预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/predict_det.py#L42)函数是否一致。算法在评估的时候，输入图像大小会影响精度，为了和论文保持一致，训练icdar15配置文件中将图像resize到[736, 1280]，但是在inference model预测的时候只有一套默认参数，会考虑到预测速度问题，默认限制图像最长边为960做resize的。训练模型预处理和inference模型的预处理函数位于[ppocr/data/imaug/operators.py](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/ppocr/data/imaug/operators.py#L147)
- 检查[trained model后处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L51)，和[inference 后处理参数](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/utility.py#L50)是否一致。
doc/doc_ch/environment.md 0 → 100644
浏览文件 @ f57a6d1b
# 运行环境准备
* [1. Python环境搭建](#1)
+ [1.1 Windows](#1.1)
+ [1.2 Mac](#1.2)
+ [1.3 Linux](#1.3)
* [2. 安装PaddlePaddle](#2)
<a name="1"></a>
## 1. Python环境搭建
<a name="1.1"></a>
### 1.1 Windows
#### 1.1.1 安装Anaconda
- 说明：使用paddlepaddle需要先安装python环境，这里我们选择python集成环境Anaconda工具包
- Anaconda是1个常用的python包管理程序
- 安装完Anaconda后，可以安装python环境，以及numpy等所需的工具包环境。
- Anaconda下载：
- 地址：https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
- 大部分win10电脑均为64位操作系统，选择x86_64版本；若电脑为32位操作系统，则选择x86.exe
<img src="../install/windows/Anaconda_download.png" alt="anaconda download" width="800" align="center"/>
- 下载完成后，双击安装程序进入图形界面
- 默认安装位置为C盘，建议将安装位置更改到D盘：
<img src="../install/windows/anaconda_install_folder.png" alt="install config" width="500" align="center"/>
- 勾选conda加入环境变量，忽略警告：
<img src="../install/windows/anaconda_install_env.png" alt="add conda to path" width="500" align="center"/>
#### 1.1.2 打开终端并创建conda环境
- 打开Anaconda Prompt终端：左下角Windows Start Menu -> Anaconda3 -> Anaconda Prompt启动控制台
<img src="../install/windows/anaconda_prompt.png" alt="anaconda download" width="300" align="center"/>
- 创建新的conda环境
```shell
# 在命令行输入以下命令，创建名为paddle_env的环境
# 此处为加速下载，使用清华源
conda create --name paddle_env python=3.8 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ # 这是一行命令
```
该命令会创建1个名为paddle_env、python版本为3.8的可执行环境，根据网络状态，需要花费一段时间
之后命令行中会输出提示信息，输入y并回车继续安装
<img src="../install/windows/conda_new_env.png" alt="conda create" width="700" align="center"/>
- 激活刚创建的conda环境，在命令行中输入以下命令：
```shell
# 激活paddle_env环境
conda activate paddle_env
# 查看当前python的位置
where python
```
<img src="../install/windows/conda_list_env.png" alt="create environment" width="600" align="center"/>
以上anaconda环境和python环境安装完毕
<a name="1.2"></a>
### 1.2 Mac
#### 1.2.1 安装Anaconda
- 说明：使用paddlepaddle需要先安装python环境，这里我们选择python集成环境Anaconda工具包
- Anaconda是1个常用的python包管理程序
- 安装完Anaconda后，可以安装python环境，以及numpy等所需的工具包环境
- Anaconda下载：
- 地址：https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
<img src="../install/mac/anaconda_start.png" alt="anaconda download" width="800" align="center"/>
- 选择最下方的`Anaconda3-2021.05-MacOSX-x86_64.pkg`下载
- 下载完成后，双击.pkg文件进入图形界面
- 按默认设置即可，安装需要花费一段时间
- 建议安装vscode或pycharm等代码编辑器
#### 1.2.2 打开终端并创建conda环境
- 打开终端
- 同时按下command键和空格键，在聚焦搜索中输入"终端"，双击进入终端
- **将conda加入环境变量**
- 加入环境变量是为了让系统能识别conda命令
- 输入以下命令，在终端中打开`~/.bash_profile`：
```shell
vim ~/.bash_profile
```
- 在`~/.bash_profile`中将conda添加为环境变量：
```shell
# 先按i进入编辑模式
# 在第一行输入：
export PATH="~/opt/anaconda3/bin:$PATH"
# 若安装时自定义了安装位置，则将~/opt/anaconda3/bin改为自定义的安装目录下的bin文件夹
```
```shell
# 修改后的~/.bash_profile文件应如下（其中xxx为用户名）：
export PATH="~/opt/anaconda3/bin:$PATH"
# >>> conda initialize >>>
# !! Contents within this block are managed by 'conda init' !!
__conda_setup="$('/Users/xxx/opt/anaconda3/bin/conda' 'shell.bash' 'hook' 2> /dev/null)"
if [ $? -eq 0 ]; then
eval "$__conda_setup"
else
if [ -f "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh" ]; then
. "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh"
else
export PATH="/Users/xxx/opt/anaconda3/bin:$PATH"
fi
fi
unset __conda_setup
# <<< conda initialize <<<
```
- 修改完成后，先按`esc`键退出编辑模式，再输入`:wq!`并回车，以保存退出
- 验证是否能识别conda命令：
- 在终端中输入`source ~/.bash_profile`以更新环境变量
- 再在终端输入`conda info --envs`，若能显示当前有base环境，则conda已加入环境变量
- 创建新的conda环境
```shell
# 在命令行输入以下命令，创建名为paddle_env的环境
# 此处为加速下载，使用清华源
conda create --name paddle_env python=3.8 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
```
- 该命令会创建1个名为paddle_env、python版本为3.8的可执行环境，根据网络状态，需要花费一段时间
- 之后命令行中会输出提示信息，输入y并回车继续安装
- <img src="../install/mac/conda_create.png" alt="conda_create" width="600" align="center"/>
- 激活刚创建的conda环境，在命令行中输入以下命令：
```shell
# 激活paddle_env环境
conda activate paddle_env
# 查看当前python的位置
where python
```
<img src="../install/mac/conda_activate.png" alt="conda_actviate" width="600" align="center"/>
以上anaconda环境和python环境安装完毕
<a name="1.3"></a>
### 1.3 Linux
Linux用户可选择Anaconda或Docker两种方式运行。如果你熟悉Docker且需要训练PaddleOCR模型，推荐使用Docker环境，PaddleOCR的开发流程均在Docker环境下运行。如果你不熟悉Docker，也可以使用Anaconda来运行项目。
#### 1.3.1 Anaconda环境配置
- 说明：使用paddlepaddle需要先安装python环境，这里我们选择python集成环境Anaconda工具包
- Anaconda是1个常用的python包管理程序
- 安装完Anaconda后，可以安装python环境，以及numpy等所需的工具包环境
- **下载Anaconda**：
- 下载地址：https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
<img src="../install/linux/anaconda_download.png" akt="anaconda download" width="800" align="center"/>
- 选择适合您操作系统的版本
- 可在终端输入`uname -m`查询系统所用的指令集
- 下载法1：本地下载，再将安装包传到linux服务器上
- 下载法2：直接使用linux命令行下载
```shell
# 首先安装wget
sudo apt-get install wget # Ubuntu
sudo yum install wget # CentOS
```
```shell
# 然后使用wget从清华源上下载
# 如要下载Anaconda3-2021.05-Linux-x86_64.sh，则下载命令如下：
wget https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/Anaconda3-2021.05-Linux-x86_64.sh
# 若您要下载其他版本，需要将最后1个/后的文件名改成您希望下载的版本
```
- 安装Anaconda：
- 在命令行输入`sh Anaconda3-2021.05-Linux-x86_64.sh`
- 若您下载的是其它版本，则将该命令的文件名替换为您下载的文件名
- 按照安装提示安装即可
- 查看许可时可输入q来退出
- **将conda加入环境变量**
- 加入环境变量是为了让系统能识别conda命令，若您在安装时已将conda加入环境变量path，则可跳过本步
- 在终端中打开`~/.bashrc`：
```shell
# 在终端中输入以下命令：
vim ~/.bashrc
```
- 在`~/.bashrc`中将conda添加为环境变量：
```shell
# 先按i进入编辑模式
# 在第一行输入：
export PATH="~/anaconda3/bin:$PATH"
# 若安装时自定义了安装位置，则将~/anaconda3/bin改为自定义的安装目录下的bin文件夹
```
```shell
# 修改后的~/.bash_profile文件应如下（其中xxx为用户名）：
export PATH="~/opt/anaconda3/bin:$PATH"
# >>> conda initialize >>>
# !! Contents within this block are managed by 'conda init' !!
__conda_setup="$('/Users/xxx/opt/anaconda3/bin/conda' 'shell.bash' 'hook' 2> /dev/null)"
if [ $? -eq 0 ]; then
eval "$__conda_setup"
else
if [ -f "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh" ]; then
. "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh"
else
export PATH="/Users/xxx/opt/anaconda3/bin:$PATH"
fi
fi
unset __conda_setup
# <<< conda initialize <<<
```
- 修改完成后，先按`esc`键退出编辑模式，再输入`:wq!`并回车，以保存退出
- 验证是否能识别conda命令：
- 在终端中输入`source ~/.bash_profile`以更新环境变量
- 再在终端输入`conda info --envs`，若能显示当前有base环境，则conda已加入环境变量
- 创建新的conda环境
```shell
# 在命令行输入以下命令，创建名为paddle_env的环境
# 此处为加速下载，使用清华源
conda create --name paddle_env python=3.8 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
```
- 该命令会创建1个名为paddle_env、python版本为3.8的可执行环境，根据网络状态，需要花费一段时间
- 之后命令行中会输出提示信息，输入y并回车继续安装
<img src="../install/linux/conda_create.png" alt="conda_create" width="500" align="center"/>
- 激活刚创建的conda环境，在命令行中输入以下命令：
```shell
# 激活paddle_env环境
conda activate paddle_env
```
以上anaconda环境和python环境安装完毕
#### 1.3.2 Docker环境配置
**注意：第一次使用这个镜像，会自动下载该镜像，请耐心等待。**
```bash
# 切换到工作目录下
cd /home/Projects
# 首次运行需创建一个docker容器，再次运行时不需要运行当前命令
# 创建一个名字为ppocr的docker容器，并将当前目录映射到容器的/paddle目录下
如果您希望在CPU环境下使用docker，使用docker而不是nvidia-docker创建docker
sudo docker run --name ppocr -v $PWD:/paddle --network=host -it paddlepaddle/paddle:latest-dev-cuda10.1-cudnn7-gcc82 /bin/bash
如果使用CUDA10，请运行以下命令创建容器，设置docker容器共享内存shm-size为64G，建议设置32G以上
sudo nvidia-docker run --name ppocr -v $PWD:/paddle --shm-size=64G --network=host -it paddlepaddle/paddle:latest-dev-cuda10.1-cudnn7-gcc82 /bin/bash
您也可以访问[DockerHub](https://hub.docker.com/r/paddlepaddle/paddle/tags/)获取与您机器适配的镜像。
# ctrl+P+Q可退出docker 容器，重新进入docker 容器使用如下命令
sudo docker container exec -it ppocr /bin/bash
```
<a name="2"></a>
## 2. 安装PaddlePaddle
- 如果您的机器安装的是CUDA9或CUDA10，请运行以下命令安装
```bash
python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
```
- 如果您的机器是CPU，请运行以下命令安装
```bash
python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
```
更多的版本需求，请参照[飞桨官网安装文档](https://www.paddlepaddle.org.cn/install/quick)中的说明进行操作。
doc/doc_ch/inference_ppocr.md 0 → 100644
浏览文件 @ f57a6d1b
# PP-OCR模型库快速推理
本文介绍针对PP-OCR模型库的Python推理引擎使用方法，内容依次为文本检测、文本识别、方向分类器以及三者串联在CPU、GPU上的预测方法。
- [1. 文本检测模型推理](#文本检测模型推理)
- [2. 文本识别模型推理](#文本识别模型推理)
- [2.1 超轻量中文识别模型推理](#超轻量中文识别模型推理)
- [2.2 多语言模型的推理](#多语言模型的推理)
- [3. 方向分类模型推理](#方向分类模型推理)
- [4. 文本检测、方向分类和文字识别串联推理](#文本检测、方向分类和文字识别串联推理)
<a name="文本检测模型推理"></a>
## 1. 文本检测模型推理
文本检测模型推理，默认使用DB模型的配置参数。超轻量中文检测模型推理，可以执行如下命令：
```
# 下载超轻量中文检测模型：
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tartar xf ch_ppocr_mobile_v2.0_det_infer.tarpython3 tools/infer/predict_det.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./ch_ppocr_mobile_v2.0_det_infer/"
```
可视化文本检测结果默认保存到`./inference_results`文件夹里面，结果文件的名称前缀为'det_res'。结果示例如下：
![](/Users/zhulingfeng01/OCR/PaddleOCR/doc/imgs_results/det_res_00018069.jpg)
通过参数`limit_type`和`det_limit_side_len`来对图片的尺寸进行限制，
`limit_type`可选参数为[`max`, `min`]，
`det_limit_size_len` 为正整数，一般设置为32 的倍数，比如960。
参数默认设置为`limit_type='max', det_limit_side_len=960`。表示网络输入图像的最长边不能超过960，
如果超过这个值，会对图像做等宽比的resize操作，确保最长边为`det_limit_side_len`。
设置为`limit_type='min', det_limit_side_len=960` 则表示限制图像的最短边为960。
如果输入图片的分辨率比较大，而且想使用更大的分辨率预测，可以设置det_limit_side_len 为想要的值，比如1216：
```
python3 tools/infer/predict_det.py --image_dir="./doc/imgs/1.jpg" --det_model_dir="./inference/det_db/" --det_limit_type=max --det_limit_side_len=1216
```
如果想使用CPU进行预测，执行命令如下
```
python3 tools/infer/predict_det.py --image_dir="./doc/imgs/1.jpg" --det_model_dir="./inference/det_db/" --use_gpu=False
```
<a name="文本识别模型推理"></a>
## 2. 文本识别模型推理
<a name="超轻量中文识别模型推理"></a>
### 2.1 超轻量中文识别模型推理
超轻量中文识别模型推理，可以执行如下命令：
```
# 下载超轻量中文识别模型：
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar
tar xf ch_ppocr_mobile_v2.0_rec_infer.tar
python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words/ch/word_4.jpg" --rec_model_dir="ch_ppocr_mobile_v2.0_rec_infer"
```
![](/Users/zhulingfeng01/OCR/PaddleOCR/doc/imgs_words/ch/word_4.jpg)
执行命令后，上面图像的预测结果（识别的文本和得分）会打印到屏幕上，示例如下：
```bash
Predicts of ./doc/imgs_words/ch/word_4.jpg:('实力活力', 0.98458153)
```
<a name="多语言模型的推理"></a>
### 2.2 多语言模型的推理
如果您需要预测的是其他语言模型，在使用inference模型预测时，需要通过`--rec_char_dict_path`指定使用的字典路径, 同时为了得到正确的可视化结果，
需要通过 `--vis_font_path` 指定可视化的字体路径，`doc/fonts/` 路径下有默认提供的小语种字体，例如韩文识别：
```
python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words/korean/1.jpg" --rec_model_dir="./your inference model" --rec_char_type="korean" --rec_char_dict_path="ppocr/utils/dict/korean_dict.txt" --vis_font_path="doc/fonts/korean.ttf"
```
![](/Users/zhulingfeng01/OCR/PaddleOCR/doc/imgs_words/korean/1.jpg)
执行命令后，上图的预测结果为：
``` text
Predicts of ./doc/imgs_words/korean/1.jpg:('바탕으로', 0.9948904)
```
<a name="方向分类模型推理"></a>
## 3. 方向分类模型推理
方向分类模型推理，可以执行如下命令：
```
# 下载超轻量中文方向分类器模型：
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar
tar xf ch_ppocr_mobile_v2.0_cls_infer.tar
python3 tools/infer/predict_cls.py --image_dir="./doc/imgs_words/ch/word_4.jpg" --cls_model_dir="ch_ppocr_mobile_v2.0_cls_infer"
```
![](/Users/zhulingfeng01/OCR/PaddleOCR/doc/imgs_words/ch/word_1.jpg)
执行命令后，上面图像的预测结果（分类的方向和得分）会打印到屏幕上，示例如下：
```
Predicts of ./doc/imgs_words/ch/word_4.jpg:['0', 0.9999982]
```
<a name="文本检测、方向分类和文字识别串联推理"></a>
## 4. 文本检测、方向分类和文字识别串联推理
以超轻量中文OCR模型推理为例，在执行预测时，需要通过参数`image_dir`指定单张图像或者图像集合的路径、参数`det_model_dir`,`cls_model_dir`和`rec_model_dir`分别指定检测，方向分类和识别的inference模型路径。参数`use_angle_cls`用于控制是否启用方向分类模型。`use_mp`表示是否使用多进程。`total_process_num`表示在使用多进程时的进程数。可视化识别结果默认保存到 ./inference_results 文件夹里面。
```shell
# 使用方向分类器
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/" --cls_model_dir="./inference/cls/" --rec_model_dir="./inference/rec_crnn/" --use_angle_cls=true
# 不使用方向分类器
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/" --use_angle_cls=false
# 使用多进程
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/" --use_angle_cls=false --use_mp=True --total_process_num=6
```
执行命令后，识别结果图像如下：
![](/Users/zhulingfeng01/OCR/PaddleOCR/doc/imgs_results/system_res_00018069.jpg)
doc/doc_ch/models_and_config.md 0 → 100644
浏览文件 @ f57a6d1b
# PP-OCR模型与配置文件
PP-OCR模型与配置文件一章主要补充一些OCR模型的基本概念、配置文件的内容与作用以便对模型后续的参数调整和训练中拥有更好的体验。
本节包含三个部分，首先在[PP-OCR模型下载](./models_list.md)中解释PP-OCR模型的类型概念，并提供所有模型的下载链接。然后在[配置文件内容与生成](./config.md)中详细说明调整PP-OCR模型所需的参数。最后的[模型库快速使用](./inference.md)是对第一节PP-OCR模型库使用方法的介绍，可以通过Python推理引擎快速利用丰富的模型库模型获得测试结果。
------
下面我们首先了解一些OCR相关的基本概念：
- [1. OCR 简要介绍](#1-ocr-----)
* [1.1 OCR 检测模型基本概念](#11-ocr---------)
* [1.2 OCR 识别模型基本概念](#12-ocr---------)
* [1.3 PP-OCR模型](#13-pp-ocr--)
<a name="1-ocr-----"></a>
## 1. OCR 简要介绍
本节简要介绍OCR检测模型、识别模型的基本概念，并介绍PaddleOCR的PP-OCR模型。
OCR（Optical Character Recognition，光学字符识别）目前是文字识别的统称，已不限于文档或书本文字识别，更包括识别自然场景下的文字，又可以称为STR（Scene Text Recognition）。
OCR文字识别一般包括两个部分，文本检测和文本识别；文本检测首先利用检测算法检测到图像中的文本行；然后检测到的文本行用识别算法去识别到具体文字。
<a name="11-ocr---------"></a>
### 1.1 OCR 检测模型基本概念
文本检测就是要定位图像中的文字区域，然后通常以边界框的形式将单词或文本行标记出来。传统的文字检测算法多是通过手工提取特征的方式，特点是速度快，简单场景效果好，但是面对自然场景，效果会大打折扣。当前多是采用深度学习方法来做。
基于深度学习的文本检测算法可以大致分为以下几类：
1. 基于目标检测的方法；一般是预测得到文本框后，通过NMS筛选得到最终文本框，多是四点文本框，对弯曲文本场景效果不理想。典型算法为EAST、Text Box等方法。
2. 基于分割的方法；将文本行当成分割目标，然后通过分割结果构建外接文本框，可以处理弯曲文本，对于文本交叉场景问题效果不理想。典型算法为DB、PSENet等方法。
3. 混合目标检测和分割的方法；
<a name="12-ocr---------"></a>
### 1.2 OCR 识别模型基本概念
OCR识别算法的输入数据一般是文本行，背景信息不多，文字占据主要部分，识别算法目前可以分为两类算法：
1. 基于CTC的方法；即识别算法的文字预测模块是基于CTC的，常用的算法组合为CNN+RNN+CTC。目前也有一些算法尝试在网络中加入transformer模块等等。
2. 基于Attention的方法；即识别算法的文字预测模块是基于Attention的，常用算法组合是CNN+RNN+Attention。
<a name="13-pp-ocr--"></a>
### 1.3 PP-OCR模型
PaddleOCR 中集成了很多OCR算法，文本检测算法有DB、EAST、SAST等等，文本识别算法有CRNN、RARE、StarNet、Rosetta、SRN等算法。
其中PaddleOCR针对中英文自然场景通用OCR，推出了PP-OCR系列模型，PP-OCR模型由DB+CRNN算法组成，利用海量中文数据训练加上模型调优方法，在中文场景上具备较高的文本检测识别能力。并且PaddleOCR推出了高精度超轻量PP-OCRv2模型，检测模型仅3M，识别模型仅8.5M，利用[PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim)的模型量化方法，可以在保持精度不降低的情况下，将检测模型压缩到0.8M，识别压缩到3M，更加适用于移动端部署场景。
doc/doc_ch/models_list.md
浏览文件 @ f57a6d1b
## OCR模型列表（V2.0，2021年1月20日更新）
## OCR模型列表（V2.1，2021年9月6日更新）
> **说明**
> 1. 2.0版模型和[1.1版模型](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/models_list.md) 的主要区别在于动态图训练vs.静态图训练，模型性能上无明显差距。
> 2. 本文档提供的是PPOCR自研模型列表，更多基于公开数据集的算法介绍与预训练模型可以参考：[算法概览文档](./algorithm_overview.md)。
> 1. 2.1版模型相比2.0版模型，2.1的模型在模型精度上做了提升
> 2. 2.0版模型和[1.1版模型](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/models_list.md) 的主要区别在于动态图训练vs.静态图训练，模型性能上无明显差距。
> 3. 本文档提供的是PPOCR自研模型列表，更多基于公开数据集的算法介绍与预训练模型可以参考：[算法概览文档](./algorithm_overview.md)。
- [一、文本检测模型](#文本检测模型)
......@@ -32,6 +33,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
|模型名称|模型简介|配置文件|推理模型大小|下载地址|
| --- | --- | --- | --- | --- |
|ch_ppocr_mobile_slim_v2.1_det|slim量化+蒸馏版超轻量模型，支持中英文、多语种文本检测|[ch_det_lite_train_cml_v2.1.yml](../../configs/det/ch_ppocr_v2.1/ch_det_lite_train_cml_v2.1.yml)| 3M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_slim_quant_infer.tar)|
|ch_ppocr_mobile_v2.1_det|原始超轻量模型，支持中英文、多语种文本检测|[ch_det_lite_train_cml_v2.1.ym](../../configs/det/ch_ppocr_v2.1/ch_det_lite_train_cml_v2.1.yml)|3M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_distill_train.tar)|
|ch_ppocr_mobile_slim_v2.0_det|slim裁剪版超轻量模型，支持中英文、多语种文本检测|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)| 2.6M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_prune_infer.tar)|
|ch_ppocr_mobile_v2.0_det|原始超轻量模型，支持中英文、多语种文本检测|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|3M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|
|ch_ppocr_server_v2.0_det|通用模型，支持中英文、多语种文本检测，比超轻量模型更大，但效果更好|[ch_det_res18_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_res18_db_v2.0.yml)|47M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)|
......@@ -45,6 +48,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
|模型名称|模型简介|配置文件|推理模型大小|下载地址|
| --- | --- | --- | --- | --- |
|ch_ppocr_mobile_slim_v2.1_rec|slim量化版超轻量模型，支持中英文、数字识别|[rec_chinese_lite_train_distillation_v2.1.yml](../../configs/rec/ch_ppocr_v2.1/rec_chinese_lite_train_distillation_v2.1.yml)| 9M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_slim_quant_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_slim_quant_train.tar) |
|ch_ppocr_mobile_v2.1_rec|原始超轻量模型，支持中英文、数字识别|[rec_chinese_lite_train_distillation_v2.1.yml](../../configs/rec/ch_ppocr_v2.1/rec_chinese_lite_train_distillation_v2.1.yml)|8.5M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_train.tar) |
|ch_ppocr_mobile_slim_v2.0_rec|slim裁剪量化版超轻量模型，支持中英文、数字识别|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)| 6M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_train.tar) |
|ch_ppocr_mobile_v2.0_rec|原始超轻量模型，支持中英文、数字识别|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)|5.2M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
|ch_ppocr_server_v2.0_rec|通用模型，支持中英文、数字识别|[rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml)|94.8M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
......@@ -62,46 +67,6 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
<a name="多语言识别模型"></a>
#### 3. 多语言识别模型（更多语言持续更新中...）
**说明：** 新增的多语言模型的配置文件通过代码方式生成，您可以通过`--help`参数查看当前PaddleOCR支持生成哪些多语言的配置文件：
```bash
# 该代码需要在指定目录运行
cd {your/path/}PaddleOCR/configs/rec/multi_language/
python3 generate_multi_language_configs.py --help
```
下面以生成意大利语配置文件为例：
##### 1. 生成意大利语配置文件测试现有模型
如果您仅仅想用配置文件测试PaddleOCR提供的多语言模型可以通过下面命令生成默认的配置文件，使用PaddleOCR提供的小语种字典进行预测。
```bash
# 该代码需要在指定目录运行
cd {your/path/}PaddleOCR/configs/rec/multi_language/
# 通过-l或者--language参数设置需要生成的语种的配置文件，该命令会将默认参数写入配置文件
python3 generate_multi_language_configs.py -l it
```
##### 2. 生成意大利语配置文件训练自己的数据
如果您想训练自己的小语种模型，可以准备好训练集文件、验证集文件、字典文件和训练数据路径，这里假设准备的意大利语的训练集、验证集、字典和训练数据路径为：
- 训练集:{your/path/}PaddleOCR/train_data/train_list.txt
- 验证集:{your/path/}PaddleOCR/train_data/val_list.txt
- 使用PaddleOCR提供的默认字典：{your/path/}PaddleOCR/ppocr/utils/dict/it_dict.txt
- 训练数据路径:{your/path/}PaddleOCR/train_data
使用以下命令生成配置文件：
```bash
# 该代码需要在指定目录运行
cd {your/path/}PaddleOCR/configs/rec/multi_language/
# -l或者--language字段是必须的
# --train修改训练集，--val修改验证集，--data_dir修改数据集目录，-o修改对应默认参数
# --dict命令改变字典路径，示例使用默认字典路径则该参数可不填
python3 generate_multi_language_configs.py -l it \
--train train_data/train_list.txt \
--val train_data/val_list.txt \
--data_dir train_data \
-o Global.use_gpu=False
```
<a name="多语言模型与配置文件"></a>
##### 3. 多语言模型与配置文件
|模型名称|字典文件|模型简介|配置文件|推理模型大小|下载地址|
| --- | --- | --- | --- |--- | --- |
| french_mobile_v2.0_rec | ppocr/utils/dict/french_dict.txt |法文识别|[rec_french_lite_train.yml](../../configs/rec/multi_language/rec_french_lite_train.yml)|2.65M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/french_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/french_mobile_v2.0_rec_train.tar) |
......@@ -125,13 +90,15 @@ python3 generate_multi_language_configs.py -l it \
|模型名称|模型简介|配置文件|推理模型大小|下载地址|
| --- | --- | --- | --- | --- |
|ch_ppocr_mobile_slim_v2.0_cls|slim量化版模型|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) |
|ch_ppocr_mobile_v2.0_cls|原始模型|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)|1.38M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |
|ch_ppocr_mobile_slim_v2.0_cls|slim量化版模型，对检测到的文本行文字角度分类|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) |
|ch_ppocr_mobile_v2.0_cls|原始分类器模型，对检测到的文本行文字角度分类|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)|1.38M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |
<a name="Paddle-Lite模型"></a>
### 四、Paddle-Lite 模型
|模型版本|模型简介|模型大小|检测模型|文本方向分类模型|识别模型|Paddle-Lite版本|
|---|---|---|---|---|---|---|
|V2.0|超轻量中文OCR 移动端模型|7.8M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
|V2.0(slim)|超轻量中文OCR 移动端模型|3.3M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_slim_opt.nb)|v2.9|
|V2.1|ppocr_v2.1蒸馏版超轻量中文OCR移动端模型|11M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_infer_opt.nb)|v2.9|
|V2.1(slim)|ppocr_v2.1蒸馏版超轻量中文OCR移动端模型|4.9M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_slim_opt.nb)|v2.9|
|V2.0|ppocr_v2.0超轻量中文OCR移动端模型|7.8M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
|V2.0(slim)|ppocr_v2.0超轻量中文OCR移动端模型|3.3M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_slim_opt.nb)|v2.9|
doc/doc_ch/multi_languages.md
浏览文件 @ f57a6d1b
......@@ -200,9 +200,9 @@ ppocr 支持使用自己的数据进行自定义训练或finetune, 其中识别
|英文|english|en| |乌克兰文|Ukranian|uk|
|法文|french|fr| |白俄罗斯文|Belarusian|be|
|德文|german|german| |泰卢固文|Telugu |te|
|日文|japan|japan| | |阿巴扎文|Abaza |abq|
|日文|japan|japan| | 阿巴扎文 | Abaza | abq |
|韩文|korean|korean| |泰米尔文|Tamil |ta|
|中文繁体|chinese traditional |ch_tra| |南非荷兰文 |Afrikaans |af|
|中文繁体|chinese traditional |chinese_cht| |南非荷兰文 |Afrikaans |af|
|意大利文| Italian |it| |阿塞拜疆文 |Azerbaijani |az|
|西班牙文|Spanish |es| |波斯尼亚文|Bosnian|bs|
|葡萄牙文| Portuguese|pt| |捷克文|Czech|cs|
......
doc/doc_ch/paddleOCR_overview.md 0 → 100644
浏览文件 @ f57a6d1b
# PaddleOCR全景图与项目克隆
## 1. PaddleOCR全景图
PaddleOCR包含丰富的文本检测、文本识别以及端到端算法。结合实际测试与产业经验，PaddleOCR选择DB和CRNN作为基础的检测和识别模型，经过一系列优化策略提出面向产业应用的PP-OCR模型。PP-OCR模型针对通用场景，根据不同语种形成了PP-OCR模型库。基于PP-OCR的能力，PaddleOCR针对文档场景任务发布PP-Structure工具库，包含版面分析和表格识别两大任务。为了打通产业落地的全流程，PaddleOCR提供了规模化的数据生产工具和多种预测部署工具，助力开发者快速落地。
<div align="center">
<img src="../overview.png">
</div>
## 2. 项目克隆
### **2.1 克隆PaddleOCR repo代码**
```
【推荐】git clone https://github.com/PaddlePaddle/PaddleOCR
```
如果因为网络问题无法pull成功，也可选择使用码云上的托管：
```
git clone https://gitee.com/paddlepaddle/PaddleOCR
```
注：码云托管代码可能无法实时同步本github项目更新，存在3~5天延时，请优先使用推荐方式。
### **2.2 安装第三方库**
```
cd PaddleOCR
pip3 install -r requirements.txt
```
doc/doc_ch/quickstart.md
浏览文件 @ f57a6d1b
# PaddleOCR快速开始
# 中文OCR模型快速使用
## 1.环境配置
- [PaddleOCR快速开始](#paddleocr)
+ [1. 安装PaddleOCR whl包](#1)
* [2. 便捷使用](#2)
+ [2.1 命令行使用](#21)
- [2.1.1 中英文模型](#211)
- [2.1.2 多语言模型](#212)
- [2.1.3 版面分析](#213)
+ [2.2 Python脚本使用](#22)
- [2.2.1 中英文与多语言使用](#221)
- [2.2.2 版面分析](#222)
请先参考[快速安装](./installation.md)配置PaddleOCR运行环境。
<a name="1"></a>
*注意：也可以通过 whl 包安装使用PaddleOCR，具体参考[Paddleocr Package使用说明](./whl.md)。*
## 1. 安装PaddleOCR whl包
## 2.inference模型下载
```bash
pip install "paddleocr>=2.0.1" # 推荐使用2.0.1+版本
```
* 移动端和服务器端的检测与识别模型如下，更多模型下载（包括多语言），可以参考[PP-OCR v2.0 系列模型下载](../doc_ch/models_list.md)
- 对于Windows环境用户：
| 模型简介 | 模型名称 |推荐场景 | 检测模型 | 方向分类器 | 识别模型 |
| ------------ | --------------- | ----------------|---- | ---------- | -------- |
| 中英文超轻量OCR模型（8.1M） | ch_ppocr_mobile_v2.0_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
| 中英文通用OCR模型（143M） | ch_ppocr_server_v2.0_xx |服务器端 |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
直接通过pip安装的shapely库可能出现`[winRrror 126] 找不到指定模块的问题`。建议从[这里](https://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely)下载shapely安装包完成安装，
- 使用**版面分析**功能时，运行以下命令**安装 Layout-Parser**
* windows 环境下如果没有安装wget,下载模型时可将链接复制到浏览器中下载，并解压放置在相应目录下
```bash
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
```
复制上表中的检测和识别的`inference模型`下载地址，并解压
```
mkdir inference && cd inference
# 下载检测模型并解压
wget {url/of/detection/inference_model} && tar xf {name/of/detection/inference_model/package}
# 下载识别模型并解压
wget {url/of/recognition/inference_model} && tar xf {name/of/recognition/inference_model/package}
# 下载方向分类器模型并解压
wget {url/of/classification/inference_model} && tar xf {name/of/classification/inference_model/package}
cd ..
```
<a name="2"></a>
## 2. 便捷使用
<a name="21"></a>
### 2.1 命令行使用
以超轻量级模型为例：
PaddleOCR提供了一系列测试图片，点击[这里](https://paddleocr.bj.bcebos.com/dygraph_v2.1/ppocr_img.zip)下载并解压，然后在终端中切换到相应目录
```
mkdir inference && cd inference
# 下载超轻量级中文OCR模型的检测模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_det_infer.tar
# 下载超轻量级中文OCR模型的识别模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar
# 下载超轻量级中文OCR模型的文本方向分类器模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar && tar xf ch_ppocr_mobile_v2.0_cls_infer.tar
cd ..
cd /path/to/ppocr_img
```
解压完毕后应有如下文件结构：
如果不使用提供的测试图片，可以将下方`--image_dir`参数替换为相应的测试图片路径
<a name="211"></a>
#### 2.1.1 中英文模型
* 检测+方向分类器+识别全流程：设置方向分类器参数`--use_angle_cls true`后可对竖排文本进行识别。
```bash
paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true
```
结果是一个list，每个item包含了文本框，文字和识别置信度
```bash
[[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]], ['纯臻营养护发素', 0.964739]]
[[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]], ['产品信息/参数', 0.98069626]]
[[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]], ['（45元/每公斤，100公斤起订）', 0.9676722]]
......
```
- 单独使用检测：设置`--rec`为`false`
```bash
paddleocr --image_dir ./imgs/11.jpg --rec false
```
结果是一个list，每个item只包含文本框
```bash
[[26.0, 457.0], [137.0, 457.0], [137.0, 477.0], [26.0, 477.0]]
[[25.0, 425.0], [372.0, 425.0], [372.0, 448.0], [25.0, 448.0]]
[[128.0, 397.0], [273.0, 397.0], [273.0, 414.0], [128.0, 414.0]]
......
```
- 单独使用识别：设置`--det`为`false`
```bash
paddleocr --image_dir ./imgs_words/ch/word_1.jpg --det false
```
结果是一个list，每个item只包含识别结果和识别置信度
```bash
['韩国小馆', 0.9907421]
```
更多whl包使用可参考[whl包文档](./whl.md)
<a name="212"></a>
#### 2.1.2 多语言模型
Paddleocr目前支持80个语种，可以通过修改`--lang`参数进行切换，对于英文模型，指定`--lang=en`。
``` bash
paddleocr --image_dir ./imgs_en/254.jpg --lang=en
```
├── ch_ppocr_mobile_v2.0_cls_infer
│ ├── inference.pdiparams
│ ├── inference.pdiparams.info
│ └── inference.pdmodel
├── ch_ppocr_mobile_v2.0_det_infer
│ ├── inference.pdiparams
│ ├── inference.pdiparams.info
│ └── inference.pdmodel
├── ch_ppocr_mobile_v2.0_rec_infer
├── inference.pdiparams
├── inference.pdiparams.info
└── inference.pdmodel
<div align="center">
<img src="../imgs_en/254.jpg" width="300" height="600">
<img src="../imgs_results/multi_lang/img_02.jpg" width="600" height="600">
</div>
结果是一个list，每个item包含了文本框，文字和识别置信度
```text
[('PHO CAPITAL', 0.95723116), [[66.0, 50.0], [327.0, 44.0], [327.0, 76.0], [67.0, 82.0]]]
[('107 State Street', 0.96311164), [[72.0, 90.0], [451.0, 84.0], [452.0, 116.0], [73.0, 121.0]]]
[('Montpelier Vermont', 0.97389287), [[69.0, 132.0], [501.0, 126.0], [501.0, 158.0], [70.0, 164.0]]]
[('8022256183', 0.99810505), [[71.0, 175.0], [363.0, 170.0], [364.0, 202.0], [72.0, 207.0]]]
[('REG 07-24-201706:59 PM', 0.93537045), [[73.0, 299.0], [653.0, 281.0], [654.0, 318.0], [74.0, 336.0]]]
[('045555', 0.99346405), [[509.0, 331.0], [651.0, 325.0], [652.0, 356.0], [511.0, 362.0]]]
[('CT1', 0.9988654), [[535.0, 367.0], [654.0, 367.0], [654.0, 406.0], [535.0, 406.0]]]
......
```
## 3.单张图像或者图像集合预测
常用的多语言简写包括
以下代码实现了文本检测、方向分类器和识别串联推理，在执行预测时，需要通过参数image_dir指定单张图像或者图像集合的路径、参数`det_model_dir`指定检测inference模型的路径、参数`rec_model_dir`指定识别inference模型的路径、参数`use_angle_cls`指定是否使用方向分类器、参数`cls_model_dir`指定方向分类器inference模型的路径、参数`use_space_char`指定是否预测空格字符。可视化识别结果默认保存到`./inference_results`文件夹里面。
| 语种 | 缩写 | | 语种 | 缩写 | | 语种 | 缩写 |
| -------- | ----------- | ---- | -------- | ------ | ---- | -------- | ------ |
| 中文 | ch | | 法文 | fr | | 日文 | japan |
| 英文 | en | | 德文 | german | | 韩文 | korean |
| 繁体中文 | chinese_cht | | 意大利文 | it | | 俄罗斯文 | ru |
```bash
全部语种及其对应的缩写列表可查看[多语言模型教程](./multi_languages.md)
<a name="213"></a>
# 预测image_dir指定的单张图像
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True
#### 2.1.3 版面分析
# 预测image_dir指定的图像集合
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True
版面分析是指对文档图片中的文字、标题、列表、图片和表格5类区域进行划分。对于前三类区域，直接使用OCR模型完成对应区域文字检测与识别，并将结果保存在txt中。对于表格类区域，经过表格结构化处理后，表格图片转换为相同表格样式的Excel文件。图片区域会被单独裁剪成图像。
# 如果想使用CPU进行预测，需设置use_gpu参数为False
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True --use_gpu=False
使用PaddleOCR的版面分析功能，需要指定`--type=structure`
```bash
paddleocr --image_dir=./table/1.png --type=structure
```
- 通用中文OCR模型
- **返回结果说明**
PP-Structure的返回结果为一个dict组成的list，示例如下
```shell
[{ 'type': 'Text',
'bbox': [34, 432, 345, 462],
'res': ([[36.0, 437.0, 341.0, 437.0, 341.0, 446.0, 36.0, 447.0], [41.0, 454.0, 125.0, 453.0, 125.0, 459.0, 41.0, 460.0]],
[('Tigure-6. The performance of CNN and IPT models using difforen', 0.90060663), ('Tent ', 0.465441)])
}
]
```
其中各个字段说明如下
| 字段 | 说明 |
| ---- | ------------------------------------------------------------ |
| type | 图片区域的类型 |
| bbox | 图片区域的在原图的坐标，分别[左上角x，左上角y，右下角x，右下角y] |
| res | 图片区域的OCR或表格识别结果。<br>表格: 表格的HTML字符串; <br>OCR: 一个包含各个单行文字的检测坐标和识别结果的元组 |
运行完成后，每张图片会在`output`字段指定的目录下有一个同名目录，图片里的每个表格会存储为一个excel，图片区域会被裁剪之后保存下来，excel文件和图片名为表格在图片里的坐标。
```
/output/table/1/
└─ res.txt
└─ [454, 360, 824, 658].xlsx 表格识别结果
└─ [16, 2, 828, 305].jpg 被裁剪出的图片区域
└─ [17, 361, 404, 711].xlsx 表格识别结果
```
- **参数说明**
| 字段 | 说明 | 默认值 |
| --------------- | ---------------------------------------- | -------------------------------------------- |
| output | excel和识别结果保存的地址 | ./output/table |
| table_max_len | 表格结构模型预测时，图像的长边resize尺度 | 488 |
| table_model_dir | 表格结构模型 inference 模型地址 | None |
| table_char_type | 表格结构模型所用字典地址 | ../ppocr/utils/dict/table_structure_dict.txt |
大部分参数和paddleocr whl包保持一致，见 [whl包文档](./whl.md)
<a name="22"></a>
### 2.2 Python脚本使用
<a name="221"></a>
#### 2.2.1 中英文与多语言使用
请按照上述步骤下载相应的模型，并且更新相关的参数，示例如下：
通过Python脚本使用PaddleOCR whl包，whl包会自动下载ppocr轻量级模型作为默认模型。
* 检测+方向分类器+识别全流程
```python
from paddleocr import PaddleOCR, draw_ocr
# Paddleocr目前支持的多语言语种可以通过修改lang参数进行切换
# 例如`ch`, `en`, `fr`, `german`, `korean`, `japan`
ocr = PaddleOCR(use_angle_cls=True, lang="ch") # need to run only once to download and load model into memory
img_path = './imgs/11.jpg'
result = ocr.ocr(img_path, cls=True)
for line in result:
print(line)
# 显示结果
from PIL import Image
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='./fonts/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
结果是一个list，每个item包含了文本框，文字和识别置信度
```bash
# 预测image_dir指定的单张图像
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_server_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_server_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True
[[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]], ['纯臻营养护发素', 0.964739]]
[[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]], ['产品信息/参数', 0.98069626]]
[[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]], ['（45元/每公斤，100公斤起订）', 0.9676722]]
......
```
* 注意：
- 如果希望使用不支持空格的识别模型，在预测的时候需要注意：请将代码更新到最新版本，并添加参数 `--use_space_char=False`。
- 如果不希望使用方向分类器，在预测的时候需要注意：请将代码更新到最新版本，并添加参数 `--use_angle_cls=False`。
结果可视化
<div align="center">
<img src="../imgs_results/whl/11_det_rec.jpg" width="800">
</div>
<a name="222"></a>
#### 2.2.2 版面分析
```python
import os
import cv2
from paddleocr import PPStructure,draw_structure_result,save_structure_res
table_engine = PPStructure(show_log=True)
更多的文本检测、识别串联推理使用方式请参考文档教程中[基于Python预测引擎推理](./inference.md)。
save_folder = './output/table'
img_path = './table/paper-image.jpg'
img = cv2.imread(img_path)
result = table_engine(img)
save_structure_res(result, save_folder,os.path.basename(img_path).split('.')[0])
此外，文档教程中也提供了中文OCR模型的其他预测部署方式：
- [基于C++预测引擎推理](../../deploy/cpp_infer/readme.md)
- [服务部署](../../deploy/hubserving)
- [端侧部署（目前只支持静态图）](https://github.com/PaddlePaddle/PaddleOCR/tree/develop/deploy/lite)
for line in result:
line.pop('img')
print(line)
from PIL import Image
font_path = './fonts/simfang.ttf' # PaddleOCR下提供字体包
image = Image.open(img_path).convert('RGB')
im_show = draw_structure_result(image, result,font_path=font_path)
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
doc/doc_ch/recognition.md
浏览文件 @ f57a6d1b
## 文字识别
# 文字识别
本文提供了PaddleOCR文本识别任务的全流程指南，包括数据准备、模型训练、调优、评估、预测，各个阶段的详细说明：
- [1 数据准备](#数据准备)
- [1.1 自定义数据集](#自定义数据集)
......@@ -9,22 +10,21 @@
- [2 启动训练](#启动训练)
- [2.1 数据增强](#数据增强)
- [2.2 训练](#训练)
- [2.3 小语种](#小语种)
- [2.2 通用模型训练](#通用模型训练)
- [2.3 多语言模型训练](#多语言模型训练)
- [3 评估](#评估)
- [4 预测](#预测)
- [4.1 训练引擎预测](#训练引擎预测)
<a name="数据准备"></a>
### 1. 数据准备
## 1. 数据准备
PaddleOCR 支持两种数据格式:
- `lmdb` 用于训练以lmdb格式存储的数据集;
- `通用数据` 用于训练以文本文件存储的数据集:
- `lmdb` 用于训练以lmdb格式存储的数据集(LMDBDataSet);
- `通用数据` 用于训练以文本文件存储的数据集(SimpleDataSet);
训练数据的默认存储路径是 `PaddleOCR/train_data`,如果您的磁盘上已有数据集，只需创建软链接至数据集目录：
......@@ -36,7 +36,7 @@ mklink /d <path/to/paddle_ocr>/train_data/dataset <path/to/dataset>
```
<a name="准备数据集"></a>
#### 1.1 自定义数据集
### 1.1 自定义数据集
下面以通用数据集为例， 介绍如何准备数据集：
* 训练集
......@@ -82,16 +82,16 @@ train_data/rec/train/word_002.jpg 用科技让复杂的世界更简单
<a name="数据下载"></a>
1.2 数据下载
### 1.2 数据下载
若您本地没有数据集，可以在官网下载 [icdar2015](http://rrc.cvc.uab.es/?ch=4&com=downloads) 数据，用于快速验证。也可以参考[DTRB](https://github.com/clovaai/deep-text-recognition-benchmark#download-lmdb-dataset-for-traininig-and-evaluation-from-here) ，下载 benchmark 所需的lmdb格式数据集。
- ICDAR2015
如果你使用的是icdar2015的公开数据集，PaddleOCR 提供了一份用于训练 icdar2015 数据集的标签文件，通过以下方式下载：
如果希望复现SRN的论文指标，需要下载离线[增广数据](https://pan.baidu.com/s/1-HSZ-ZVdqBF2HaBZ5pRAKA),提取码: y3ry。增广数据是由MJSynth和SynthText做旋转和扰动得到的。数据下载完成后请解压到 {your_path}/PaddleOCR/train_data/data_lmdb_release/training/ 路径下。
若您本地没有数据集，可以在官网下载 [ICDAR2015](http://rrc.cvc.uab.es/?ch=4&com=downloads) 数据，用于快速验证。也可以参考[DTRB](https://github.com/clovaai/deep-text-recognition-benchmark#download-lmdb-dataset-for-traininig-and-evaluation-from-here) ，下载 benchmark 所需的lmdb格式数据集。
如果希望复现SAR的论文指标，需要下载[SynthAdd](https://pan.baidu.com/share/init?surl=uV0LtoNmcxbO-0YA7Ch4dg), 提取码：627x。此外，真实数据集icdar2013, icdar2015, cocotext, IIIT5也作为训练数据的一部分。具体数据细节可以参考论文SAR。
如果你使用的是icdar2015的公开数据集，PaddleOCR 提供了一份用于训练 ICDAR2015 数据集的标签文件，通过以下方式下载：
```
# 训练集标签
wget -P ./train_data/ic15_data https://paddleocr.bj.bcebos.com/dataset/rec_gt_train.txt
......@@ -99,15 +99,25 @@ wget -P ./train_data/ic15_data https://paddleocr.bj.bcebos.com/dataset/rec_gt_t
wget -P ./train_data/ic15_data https://paddleocr.bj.bcebos.com/dataset/rec_gt_test.txt
```
PaddleOCR 也提供了数据格式转换脚本，可以将官网 label 转换支持的数据格式。 数据转换工具在 `ppocr/utils/gen_label.py`, 这里以训练集为例：
PaddleOCR 也提供了数据格式转换脚本，可以将ICDAR官网 label 转换为PaddleOCR支持的数据格式。 数据转换工具在 `ppocr/utils/gen_label.py`, 这里以训练集为例：
```
# 将官网下载的标签文件转换为 rec_gt_label.txt
python gen_label.py --mode="rec" --input_path="{path/of/origin/label}" --output_label="rec_gt_label.txt"
```
数据样式格式如下，(a)为原始图片,(b)为每张图片对应的 Ground Truth 文本文件：
![](../datasets/icdar_rec.png)
- 多语言数据集
多语言模型的训练数据集均为100w的合成数据，使用了开源合成工具 [text_renderer](https://github.com/Sanster/text_renderer) ，少量的字体可以通过下面两种方式下载。
* [百度网盘](https://pan.baidu.com/s/1bS_u207Rm7YbY33wOECKDA) 提取码：frgi
* [google drive](https://drive.google.com/file/d/18cSWX7wXSy4G0tbKJ0d9PuIaiwRLHpjA/view)
<a name="字典"></a>
1.3 字典
### 1.3 字典
最后需要提供一个字典（{word_dict_name}.txt），使模型在训练时，可以将所有出现的字符映射为字典的索引。
......@@ -154,13 +164,27 @@ PaddleOCR内置了一部分字典，可以按需使用。
并将 `character_type` 设置为 `ch`。
<a name="支持空格"></a>
1.4 添加空格类别
### 1.4 添加空格类别
如果希望支持识别"空格"类别, 请将yml文件中的 `use_space_char` 字段设置为 `True`。
<a name="启动训练"></a>
### 2. 启动训练
## 2. 启动训练
<a name="数据增强"></a>
### 2.1 数据增强
PaddleOCR提供了多种数据增强方式，默认配置文件中已经添加了数据增广。
默认的扰动方式有：颜色空间转换(cvtColor)、模糊(blur)、抖动(jitter)、噪声(Gasuss noise)、随机切割(random crop)、透视(perspective)、颜色反转(reverse)、TIA数据增广。
训练过程中每种扰动方式以40%的概率被选择，具体代码实现请参考：[rec_img_aug.py](../../ppocr/data/imaug/rec_img_aug.py)
*由于OpenCV的兼容性问题，扰动操作暂时只支持Linux*
<a name="通用模型训练"></a>
### 2.2 通用模型训练
PaddleOCR提供了训练脚本、评估脚本和预测脚本，本节将以 CRNN 识别模型为例：
......@@ -180,23 +204,16 @@ tar -xf rec_mv3_none_bilstm_ctc_v2.0_train.tar && rm -rf rec_mv3_none_bilstm_ctc
*如果您安装的是cpu版本，请将配置文件中的 `use_gpu` 字段修改为false*
```
# GPU训练 支持单卡，多卡训练，通过--gpus参数指定卡号
# GPU训练 支持单卡，多卡训练
# 训练icdar15英文数据 训练日志会自动保存为 "{save_model_dir}" 下的train.log
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/rec/rec_icdar15_train.yml
```
<a name="数据增强"></a>
#### 2.1 数据增强
PaddleOCR提供了多种数据增强方式，默认配置文件中已经添加了数据增广。
默认的扰动方式有：颜色空间转换(cvtColor)、模糊(blur)、抖动(jitter)、噪声(Gasuss noise)、随机切割(random crop)、透视(perspective)、颜色反转(reverse)、TIA数据增广。
训练过程中每种扰动方式以40%的概率被选择，具体代码实现请参考：[rec_img_aug.py](../../ppocr/data/imaug/rec_img_aug.py)
#单卡训练（训练周期长，不建议）
python3 tools/train.py -c configs/rec/rec_icdar15_train.yml
*由于OpenCV的兼容性问题，扰动操作暂时只支持Linux*
#多卡训练，通过--gpus参数指定卡号
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/rec/rec_icdar15_train.yml
```
<a name="训练"></a>
#### 2.2 训练
PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_train.yml` 中修改 `eval_batch_step` 设置评估频率，默认每500个iter评估一次。评估过程中默认将最佳acc模型，保存为 `output/rec_CRNN/best_accuracy` 。
......@@ -286,85 +303,12 @@ Eval:
```
**注意，预测/评估时的配置文件请务必与训练一致。**
<a name="小语种"></a>
#### 2.3 小语种
<a name="多语言模型训练"></a>
### 2.3 多语言模型训练
PaddleOCR目前已支持80种（除中文外）语种识别，`configs/rec/multi_languages` 路径下提供了一个多语言的配置文件模版: [rec_multi_language_lite_train.yml](../../configs/rec/multi_language/rec_multi_language_lite_train.yml)。
您有两种方式创建所需的配置文件：
1. 通过脚本自动生成
[generate_multi_language_configs.py](../../configs/rec/multi_language/generate_multi_language_configs.py) 可以帮助您生成多语言模型的配置文件
- 以意大利语为例，如果您的数据是按如下格式准备的：
```
|-train_data
|- it_train.txt # 训练集标签
|- it_val.txt # 验证集标签
|- data
|- word_001.jpg
|- word_002.jpg
|- word_003.jpg
| ...
```
可以使用默认参数，生成配置文件：
```bash
# 该代码需要在指定目录运行
cd PaddleOCR/configs/rec/multi_language/
# 通过-l或者--language参数设置需要生成的语种的配置文件，该命令会将默认参数写入配置文件
python3 generate_multi_language_configs.py -l it
```
- 如果您的数据放置在其他位置，或希望使用自己的字典，可以通过指定相关参数来生成配置文件:
```bash
# -l或者--language字段是必须的
# --train修改训练集，--val修改验证集，--data_dir修改数据集目录，--dict修改字典路径， -o修改对应默认参数
cd PaddleOCR/configs/rec/multi_language/
python3 generate_multi_language_configs.py -l it \ # 语种
--train {path/of/train_label.txt} \ # 训练标签文件的路径
--val {path/of/val_label.txt} \ # 验证集标签文件的路径
--data_dir {train_data/path} \ # 训练数据的根目录
--dict {path/of/dict} \ # 字典文件路径
-o Global.use_gpu=False # 是否使用gpu
...
```
意大利文由拉丁字母组成，因此执行完命令后会得到名为 rec_latin_lite_train.yml 的配置文件。
2. 手动修改配置文件
您也可以手动修改模版中的以下几个字段:
```
Global:
use_gpu: True
epoch_num: 500
...
character_type: it # 需要识别的语种
character_dict_path: {path/of/dict} # 字典文件所在路径
Train:
dataset:
name: SimpleDataSet
data_dir: train_data/ # 数据存放根目录
label_file_list: ["./train_data/train_list.txt"] # 训练集label路径
...
Eval:
dataset:
name: SimpleDataSet
data_dir: train_data/ # 数据存放根目录
label_file_list: ["./train_data/val_list.txt"] # 验证集label路径
...
```
目前PaddleOCR支持的多语言算法有：
按语系划分，目前PaddleOCR支持的语种有：
| 配置文件 | 算法名称 | backbone | trans | seq | pred | language | character_type |
| :--------: | :-------: | :-------: | :-------: | :-----: | :-----: | :-----: | :-----: |
......@@ -381,10 +325,6 @@ PaddleOCR目前已支持80种（除中文外）语种识别，`configs/rec/multi
更多支持语种请参考: [多语言模型](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/multi_languages.md#%E8%AF%AD%E7%A7%8D%E7%BC%A9%E5%86%99)
多语言模型训练方式与中文模型一致，训练数据集均为100w的合成数据，少量的字体可以通过下面两种方式下载。
* [百度网盘](https://pan.baidu.com/s/1bS_u207Rm7YbY33wOECKDA)。提取码：frgi。
* [google drive](https://drive.google.com/file/d/18cSWX7wXSy4G0tbKJ0d9PuIaiwRLHpjA/view)
如您希望在现有模型效果的基础上调优，请参考下列说明修改配置文件：
以 `rec_french_lite_train` 为例：
......@@ -420,7 +360,7 @@ Eval:
...
```
<a name="评估"></a>
### 3 评估
## 3 评估
评估数据集可以通过 `configs/rec/rec_icdar15_train.yml` 修改Eval中的 `label_file_path` 设置。
......@@ -430,14 +370,29 @@ python3 -m paddle.distributed.launch --gpus '0' tools/eval.py -c configs/rec/rec
```
<a name="预测"></a>
### 4 预测
<a name="训练引擎预测"></a>
#### 4.1 训练引擎的预测
## 4 预测
使用 PaddleOCR 训练好的模型，可以通过以下脚本进行快速预测。
默认预测图片存储在 `infer_img` 里，通过 `-o Global.checkpoints` 指定权重：
默认预测图片存储在 `infer_img` 里，通过 `-o Global.checkpoints` 加载训练好的参数文件：
根据配置文件中设置的的 `save_model_dir` 和 `save_epoch_step` 字段，会有以下几种参数被保存下来：
```
output/rec/
├── best_accuracy.pdopt
├── best_accuracy.pdparams
├── best_accuracy.states
├── config.yml
├── iter_epoch_3.pdopt
├── iter_epoch_3.pdparams
├── iter_epoch_3.states
├── latest.pdopt
├── latest.pdparams
├── latest.states
└── train.log
```
其中 best_accuracy.* 是评估集上的最优模型；iter_epoch_x.* 是以 `save_epoch_step` 为间隔保存下来的模型；latest.* 是最后一个epoch的模型。
```
# 预测英文结果
......
doc/doc_ch/training.md 0 → 100644
浏览文件 @ f57a6d1b
# 模型训练
本文将介绍模型训练时需掌握的基本概念，和训练时的调优方法。
同时会简单介绍PaddleOCR模型训练数据的组成部分，以及如何在垂类场景中准备数据finetune模型。
- [1. 基本概念](#基本概念)
* [1.1 学习率](#学习率)
* [1.2 正则化](#正则化)
* [1.3 评估指标](#评估指标)
- [2. 数据与垂类场景](#数据与垂类场景)
* [2.1 训练数据](#训练数据)
* [2.2 垂类场景](#垂类场景)
* [2.3 自己构建数据集](#自己构建数据集)
* [3. 常见问题](#常见问题)
<a name="基本概念"></a>
## 1. 基本概念
OCR(Optical Character Recognition,光学字符识别)是指对图像进行分析识别处理，获取文字和版面信息的过程，是典型的计算机视觉任务，
通常由文本检测和文本识别两个子任务构成。
模型调优时需要关注以下参数：
<a name="学习率"></a>
### 1.1 学习率
学习率是训练神经网络的重要超参数之一，它代表在每一次迭代中梯度向损失函数最优解移动的步长。
在PaddleOCR中提供了多种学习率更新策略,可以通过配置文件修改，例如：
```
Optimizer:
...
lr:
name: Piecewise
decay_epochs : [700, 800]
values : [0.001, 0.0001]
warmup_epoch: 5
```
Piecewise 代表分段常数衰减，在不同的学习阶段指定不同的学习率，在每段内学习率相同。
warmup_epoch 代表在前5个epoch中，学习率将逐渐从0增加到base_lr。全部策略可以参考代码[learning_rate.py](../../ppocr/optimizer/learning_rate.py) 。
<a name="正则化"></a>
### 1.2 正则化
正则化可以有效的避免算法过拟合，PaddleOCR中提供了L1、L2正则方法，L1 和 L2 正则化是最常用的正则化方法。L1 正则化向目标函数添加正则化项，以减少参数的绝对值总和；而 L2 正则化中，添加正则化项的目的在于减少参数平方的总和。配置方法如下：
```
Optimizer:
...
regularizer:
name: L2
factor: 2.0e-05
```
<a name="评估指标"></a>
### 1.3 评估指标
（1）检测阶段：先按照检测框和标注框的IOU评估，IOU大于某个阈值判断为检测准确。这里检测框和标注框不同于一般的通用目标检测框，是采用多边形进行表示。检测准确率：正确的检测框个数在全部检测框的占比，主要是判断检测指标。检测召回率：正确的检测框个数在全部标注框的占比，主要是判断漏检的指标。
（2）识别阶段： 字符识别准确率，即正确识别的文本行占标注的文本行数量的比例，只有整行文本识别对才算正确识别。
（3）端到端统计： 端对端召回率：准确检测并正确识别文本行在全部标注文本行的占比； 端到端准确率：准确检测并正确识别文本行在 检测到的文本行数量 的占比； 准确检测的标准是检测框与标注框的IOU大于某个阈值，正确识别的的检测框中的文本与标注的文本相同。
<a name="数据与垂类场景"></a>
## 2. 数据与垂类场景
<a name="训练数据"></a>
### 2.1 训练数据
目前开源的模型，数据集和量级如下：
- 检测：
- 英文数据集，ICDAR2015
- 中文数据集，LSVT街景数据集训练数据3w张图片
- 识别：
- 英文数据集，MJSynth和SynthText合成数据，数据量上千万。
- 中文数据集，LSVT街景数据集根据真值将图crop出来，并进行位置校准，总共30w张图像。此外基于LSVT的语料，合成数据500w。
- 小语种数据集，使用不同语料和字体，分别生成了100w合成数据集，并使用ICDAR-MLT作为验证集。
其中，公开数据集都是开源的，用户可自行搜索下载，也可参考[中文数据集](./datasets.md)，合成数据暂不开源，用户可使用开源合成工具自行合成，可参考的合成工具包括[text_renderer](https://github.com/Sanster/text_renderer) 、[SynthText](https://github.com/ankush-me/SynthText) 、[TextRecognitionDataGenerator](https://github.com/Belval/TextRecognitionDataGenerator) 等。
<a name="垂类场景"></a>
### 2.2 垂类场景
PaddleOCR主要聚焦通用OCR，如果有垂类需求，您可以用PaddleOCR+垂类数据自己训练；
如果缺少带标注的数据，或者不想投入研发成本，建议直接调用开放的API，开放的API覆盖了目前比较常见的一些垂类。
<a name="自己构建数据集"></a>
### 2.3 自己构建数据集
在构建数据集时有几个经验可供参考：
（1） 训练集的数据量：
a. 检测需要的数据相对较少，在PaddleOCR模型的基础上进行Fine-tune，一般需要500张可达到不错的效果。
b. 识别分英文和中文，一般英文场景需要几十万数据可达到不错的效果，中文则需要几百万甚至更多。
（2）当训练数据量少时，可以尝试以下三种方式获取更多的数据：
a. 人工采集更多的训练数据，最直接也是最有效的方式。
b. 基于PIL和opencv基本图像处理或者变换。例如PIL中ImageFont, Image, ImageDraw三个模块将文字写到背景中，opencv的旋转仿射变换，高斯滤波等。
c. 利用数据生成算法合成数据，例如pix2pix或StyleText等算法。
<a name="常见问题"></a>
## 3. 常见问题
**Q**：训练CRNN识别时，如何选择合适的网络输入shape？
A：一般高度采用32，最长宽度的选择，有两种方法：
（1）统计训练样本图像的宽高比分布。最大宽高比的选取考虑满足80%的训练样本。
（2）统计训练样本文字数目。最长字符数目的选取考虑满足80%的训练样本。然后中文字符长宽比近似认为是1，英文认为3：1，预估一个最长宽度。
**Q**：识别训练时，训练集精度已经到达90了，但验证集精度一直在70，涨不上去怎么办？
A：训练集精度90，测试集70多的话，应该是过拟合了，有两个可尝试的方法：
（1）加入更多的增广方式或者调大增广prob的[概率](https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/ppocr/data/imaug/rec_img_aug.py#L341)，默认为0.4。
（2）调大系统的[l2 dcay值](https://github.com/PaddlePaddle/PaddleOCR/blob/a501603d54ff5513fc4fc760319472e59da25424/configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml#L47)
**Q**: 识别模型训练时，loss能正常下降，但acc一直为0
A：识别模型训练初期acc为0是正常的，多训一段时间指标就上来了。
doc/doc_ch/update.md
浏览文件 @ f57a6d1b
# 更新
- 2021.9.7 发布PaddleOCR v2.3，发布[PP-OCRv2](#PP-OCRv2)，CPU推理速度相比于PP-OCR server提升220%；效果相比于PP-OCR mobile 提升7%。
- 2021.8.3 发布PaddleOCR v2.2，新增文档结构分析[PP-Structure](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/ppstructure/README_ch.md)工具包，支持版面分析与表格识别（含Excel导出）。
- 2021.6.29 [FAQ](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/FAQ.md)新增5个高频问题，总数248个，每周一都会更新，欢迎大家持续关注。
- 2021.4.8 release 2.1版本，新增AAAI 2021论文[端到端识别算法PGNet](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/pgnet.md)开源，[多语言模型](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/multi_languages.md)支持种类增加到80+。
- 2020.12.15 更新数据合成工具[Style-Text](../../StyleText/README_ch.md)，可以批量合成大量与目标场景类似的图像，在多个场景验证，效果明显提升。
- 2020.12.07 [FAQ](../../doc/doc_ch/FAQ.md)新增5个高频问题，总数124个，并且计划以后每周一都会更新，欢迎大家持续关注。
- 2020.11.25 更新半自动标注工具[PPOCRLabel](../../PPOCRLabel/README_ch.md)，辅助开发者高效完成标注任务，输出格式与PP-OCR训练任务完美衔接。
......
doc/doc_ch/visualization.md
浏览文件 @ f57a6d1b
# 效果展示
<a name="超轻量PP-OCRv2效果展示"></a>
## 超轻量PP-OCRv2效果展示
<img src="../imgs_results/PP-OCRv2/PP-OCRv2-pic001.jpg" width="800">
<img src="../imgs_results/PP-OCRv2/PP-OCRv2-pic002.jpg" width="800">
<img src="../imgs_results/PP-OCRv2/PP-OCRv2-pic003.jpg" width="800">
<a name="超轻量ppocr_server_2.0效果展示"></a>
## 通用ppocr_server_2.0 效果展示
## 通用PP-OCR server 效果展示
<div align="center">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00006737.jpg" width="800">
......@@ -10,8 +16,6 @@
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00018069.jpg" width="800">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00056221.jpg" width="800">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00057937.jpg" width="800">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00059985.jpg" width="800">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00111002.jpg" width="800">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00077949.jpg" width="800">
<img src="../imgs_results/ch_ppocr_mobile_v2.0/00207393.jpg" width="800">
</div>
......
doc/doc_ch/whl.md
浏览文件 @ f57a6d1b
......@@ -210,7 +210,7 @@ paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --use_angle_cls true
```bash
[[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]], ['纯臻营养护发素', 0.964739]]
[[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]], ['产品信息/参数', 0.98069626]]
[[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]], ['（45元/每公斤，100公斤起订）', 0.9676722]]
[[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]], ['（45元/每公斤，100公斤起订）', 0.9676722]]µ
......
```
......
doc/doc_en/config_en.md
浏览文件 @ f57a6d1b
......@@ -51,7 +51,7 @@ Take rec_chinese_lite_train_v2.0.yml as an example
### Architecture ([ppocr/modeling](../../ppocr/modeling))
In ppocr, the network is divided into four stages: Transform, Backbone, Neck and Head
In PaddleOCR, the network is divided into four stages: Transform, Backbone, Neck and Head
| Parameter | Use | Defaults | Note |
| :---------------------: | :---------------------: | :--------------: | :--------------------: |
......@@ -120,3 +120,108 @@ In ppocr, the network is divided into four stages: Transform, Backbone, Neck and
| batch_size_per_card | Single card batch size during training | 256 | \ |
| drop_last | Whether to discard the last incomplete mini-batch because the number of samples in the data set cannot be divisible by batch_size | True | \ |
| num_workers | The number of sub-processes used to load data, if it is 0, the sub-process is not started, and the data is loaded in the main process | 8 | \ |
## 3. MULTILINGUAL CONFIG FILE GENERATION
PaddleOCR currently supports 80 (except Chinese) language recognition. A multi-language configuration file template is
provided under the path `configs/rec/multi_languages`: [rec_multi_language_lite_train.yml](../../configs/rec/multi_language/rec_multi_language_lite_train.yml)。
There are two ways to create the required configuration file:：
1. Automatically generated by script
[generate_multi_language_configs.py](../../configs/rec/multi_language/generate_multi_language_configs.py) Can help you generate configuration files for multi-language models
- Take Italian as an example, if your data is prepared in the following format:
```
|-train_data
|- it_train.txt # train_set label
|- it_val.txt # val_set label
|- data
|- word_001.jpg
|- word_002.jpg
|- word_003.jpg
| ...
```
You can use the default parameters to generate a configuration file:
```bash
# The code needs to be run in the specified directory
cd PaddleOCR/configs/rec/multi_language/
# Set the configuration file of the language to be generated through the -l or --language parameter.
# This command will write the default parameters into the configuration file
python3 generate_multi_language_configs.py -l it
```
- If your data is placed in another location, or you want to use your own dictionary, you can generate the configuration file by specifying the relevant parameters:
```bash
# -l or --language field is required
# --train to modify the training set
# --val to modify the validation set
# --data_dir to modify the data set directory
# --dict to modify the dict path
# -o to modify the corresponding default parameters
cd PaddleOCR/configs/rec/multi_language/
python3 generate_multi_language_configs.py -l it \ # language
--train {path/of/train_label.txt} \ # path of train_label
--val {path/of/val_label.txt} \ # path of val_label
--data_dir {train_data/path} \ # root directory of training data
--dict {path/of/dict} \ # path of dict
-o Global.use_gpu=False # whether to use gpu
...
```
Italian is made up of Latin letters, so after executing the command, you will get the rec_latin_lite_train.yml.
2. Manually modify the configuration file
You can also manually modify the following fields in the template:
```
Global:
use_gpu: True
epoch_num: 500
...
character_type: it # language
character_dict_path: {path/of/dict} # path of dict
Train:
dataset:
name: SimpleDataSet
data_dir: train_data/ # root directory of training data
label_file_list: ["./train_data/train_list.txt"] # train label path
...
Eval:
dataset:
name: SimpleDataSet
data_dir: train_data/ # root directory of val data
label_file_list: ["./train_data/val_list.txt"] # val label path
...
```
Currently, the multi-language algorithms supported by PaddleOCR are:
| Configuration file | Algorithm name | backbone | trans | seq | pred | language | character_type |
| :--------: | :-------: | :-------: | :-------: | :-----: | :-----: | :-----: | :-----: |
| rec_chinese_cht_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | chinese traditional | chinese_cht|
| rec_en_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | English(Case sensitive) | EN |
| rec_french_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | French | french |
| rec_ger_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | German | german |
| rec_japan_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | Japanese | japan |
| rec_korean_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | Korean | korean |
| rec_latin_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | Latin | latin |
| rec_arabic_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | arabic | ar |
| rec_cyrillic_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | cyrillic | cyrillic |
| rec_devanagari_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | devanagari | devanagari |
For more supported languages, please refer to : [Multi-language model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md#4-support-languages-and-abbreviations)
The multi-language model training method is the same as the Chinese model. The training data set is 100w synthetic data. A small amount of fonts and test data can be downloaded using the following two methods.
* [Baidu Netdisk](https://pan.baidu.com/s/1bS_u207Rm7YbY33wOECKDA),Extraction code:frgi.
* [Google drive](https://drive.google.com/file/d/18cSWX7wXSy4G0tbKJ0d9PuIaiwRLHpjA/view)
doc/doc_en/detection_en.md
浏览文件 @ f57a6d1b
# TEXT DETECTION
# CONTENT
- [Paste Your Document In Here](#paste-your-document-in-here)
- [1. TEXT DETECTION](#1-text-detection)
* [1.1 DATA PREPARATION](#11-data-preparation)
* [1.2 DOWNLOAD PRETRAINED MODEL](#12-download-pretrained-model)
* [1.3 START TRAINING](#13-start-training)
* [1.4 LOAD TRAINED MODEL AND CONTINUE TRAINING](#14-load-trained-model-and-continue-training)
* [1.5 TRAINING WITH NEW BACKBONE](#15-training-with-new-backbone)
* [1.6 EVALUATION](#16-evaluation)
* [1.7 TEST](#17-test)
* [1.8 INFERENCE MODEL PREDICTION](#18-inference-model-prediction)
- [2. FAQ](#2-faq)
# 1. TEXT DETECTION
This section uses the icdar2015 dataset as an example to introduce the training, evaluation, and testing of the detection model in PaddleOCR.
## DATA PREPARATION
## 1.1 DATA PREPARATION
The icdar2015 dataset can be obtained from [official website](https://rrc.cvc.uab.es/?ch=4&com=downloads). Registration is required for downloading.
After registering and logging in, download the part marked in the red box in the figure below. And, the content downloaded by `Training Set Images` should be saved as the folder `icdar_c4_train_imgs`, and the content downloaded by `Test Set Images` is saved as the folder `ch4_test_images`
<p align="center">
<img src="./doc/datasets/ic15_location_download.png" align="middle" width = "600"/>
<p align="center">
Decompress the downloaded dataset to the working directory, assuming it is decompressed under PaddleOCR/train_data/. In addition, PaddleOCR organizes many scattered annotation files into two separate annotation files for train and test respectively, which can be downloaded by wget:
```shell
# Under the PaddleOCR path
......@@ -36,10 +58,11 @@ The `points` in the dictionary represent the coordinates (x, y) of the four poin
If you want to train PaddleOCR on other datasets, please build the annotation file according to the above format.
## TRAINING
## 1.2 DOWNLOAD PRETRAINED MODEL
First download the pretrained model. The detection model of PaddleOCR currently supports 3 backbones, namely MobileNetV3, ResNet18_vd and ResNet50_vd. You can use the model in [PaddleClas](https://github.com/PaddlePaddle/PaddleClas/tree/release/2.0/ppcls/modeling/architectures) to replace backbone according to your needs.
And the responding download link of backbone pretrain weights can be found in (https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.0/README_cn.md#resnet%E5%8F%8A%E5%85%B6vd%E7%B3%BB%E5%88%97).
First download the pretrained model. The detection model of PaddleOCR currently supports 3 backbones, namely MobileNetV3, ResNet18_vd and ResNet50_vd. You can use the model in [PaddleClas](https://github.com/PaddlePaddle/PaddleClas/tree/develop/ppcls/modeling/architectures) to replace backbone according to your needs.
And the responding download link of backbone pretrain weights can be found in [PaddleClas repo](https://github.com/PaddlePaddle/PaddleClas#mobile-series).
```shell
cd PaddleOCR/
# Download the pre-trained model of MobileNetV3
......@@ -49,11 +72,13 @@ wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dyg
# or, download the pre-trained model of ResNet50_vd
wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams
```
#### START TRAINING
## 1.3 START TRAINING
*If CPU version installed, please set the parameter `use_gpu` to `false` in the configuration.*
```shell
python3 tools/train.py -c configs/det/det_mv3_db.yml
python3 tools/train.py -c configs/det/det_mv3_db.yml \
-o Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained
```
In the above instruction, use `-c` to select the training to use the `configs/det/det_db_mv3.yml` configuration file.
......@@ -62,16 +87,17 @@ For a detailed explanation of the configuration file, please refer to [config](.
You can also use `-o` to change the training parameters without modifying the yml file. For example, adjust the training learning rate to 0.0001
```shell
# single GPU training
python3 tools/train.py -c configs/det/det_mv3_db.yml -o Optimizer.base_lr=0.0001
python3 tools/train.py -c configs/det/det_mv3_db.yml -o \
Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained \
Optimizer.base_lr=0.0001
# multi-GPU training
# Set the GPU ID used by the '--gpus' parameter.
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/det/det_mv3_db.yml -o Optimizer.base_lr=0.0001
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/det/det_mv3_db.yml -o Global.pretrain_weights=./pretrain_models/MobileNetV3_large_x0_5_pretrained
```
#### load trained model and continue training
## 1.4 LOAD TRAINED MODEL AND CONTINUE TRAINING
If you expect to load trained model and continue the training again, you can specify the parameter `Global.checkpoints` as the model path to be loaded.
For example:
......@@ -82,9 +108,59 @@ python3 tools/train.py -c configs/det/det_mv3_db.yml -o Global.checkpoints=./you
**Note**: The priority of `Global.checkpoints` is higher than that of `Global.pretrain_weights`, that is, when two parameters are specified at the same time, the model specified by `Global.checkpoints` will be loaded first. If the model path specified by `Global.checkpoints` is wrong, the one specified by `Global.pretrain_weights` will be loaded.
## EVALUATION
## 1.5 TRAINING WITH NEW BACKBONE
The network part completes the construction of the network, and PaddleOCR divides the network into four parts, which are under [ppocr/modeling](../../ppocr/modeling). The data entering the network will pass through these four parts in sequence(transforms->backbones->
necks->heads).
```bash
├── architectures # Code for building network
├── transforms # Image Transformation Module
├── backbones # Feature extraction module
├── necks # Feature enhancement module
└── heads # Output module
```
If the Backbone to be replaced has a corresponding implementation in PaddleOCR, you can directly modify the parameters in the `Backbone` part of the configuration yml file.
However, if you want to use a new Backbone, an example of replacing the backbones is as follows:
1. Create a new file under the [ppocr/modeling/backbones](../../ppocr/modeling/backbones) folder, such as my_backbone.py.
2. Add code in the my_backbone.py file, the sample code is as follows:
```python
import paddle
import paddle.nn as nn
import paddle.nn.functional as F
class MyBackbone(nn.Layer):
def __init__(self, *args, **kwargs):
super(MyBackbone, self).__init__()
# your init code
self.conv = nn.xxxx
def forward(self, inputs):
# your network forward
y = self.conv(inputs)
return y
```
3. Import the added module in the [ppocr/modeling/backbones/\__init\__.py](../../ppocr/modeling/backbones/__init__.py) file.
After adding the four-part modules of the network, you only need to configure them in the configuration file to use, such as:
```yaml
Backbone:
name: MyBackbone
args1: args1
```
**NOTE**: More details about replace Backbone and other mudule can be found in [doc](add_new_algorithm_en.md).
## 1.6 EVALUATION
PaddleOCR calculates three indicators for evaluating performance of OCR detection task: Precision, Recall, and Hmean.
PaddleOCR calculates three indicators for evaluating performance of OCR detection task: Precision, Recall, and Hmean(F-Score).
Run the following code to calculate the evaluation indicators. The result will be saved in the test result file specified by `save_res_path` in the configuration file `det_db_mv3.yml`
......@@ -95,10 +171,9 @@ The model parameters during training are saved in the `Global.save_model_dir` di
python3 tools/eval.py -c configs/det/det_mv3_db.yml -o Global.checkpoints="{path/to/weights}/best_accuracy" PostProcess.box_thresh=0.6 PostProcess.unclip_ratio=1.5
```
* Note: `box_thresh` and `unclip_ratio` are parameters required for DB post-processing, and not need to be set when evaluating the EAST and SAST model.
* Note: `box_thresh` and `unclip_ratio` are parameters required for DB post-processing, and not need to be set when evaluating the EAST model.
## TEST
## 1.7 TEST
Test the detection result on a single image:
```shell
......@@ -107,7 +182,7 @@ python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./
When testing the DB model, adjust the post-processing threshold:
```shell
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/img_10.jpg" Global.pretrained_model="./output/det_db/best_accuracy" PostProcess.box_thresh=0.6 PostProcess.unclip_ratio=1.5
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/img_10.jpg" Global.pretrained_model="./output/det_db/best_accuracy" PostProcess.box_thresh=0.6 PostProcess.unclip_ratio=2.0
```
......@@ -115,3 +190,33 @@ Test the detection result on all images in the folder:
```shell
python3 tools/infer_det.py -c configs/det/det_mv3_db.yml -o Global.infer_img="./doc/imgs_en/" Global.pretrained_model="./output/det_db/best_accuracy"
```
## 1.8 INFERENCE MODEL PREDICTION
The inference model (the model saved by `paddle.jit.save`) is generally a solidified model saved after the model training is completed, and is mostly used to give prediction in deployment.
The model saved during the training process is the checkpoints model, which saves the parameters of the model and is mostly used to resume training.
Compared with the checkpoints model, the inference model will additionally save the structural information of the model. Therefore, it is easier to deploy because the model structure and model parameters are already solidified in the inference model file, and is suitable for integration with actual systems.
Firstly, we can convert DB trained model to inference model:
```shell
python3 tools/export_model.py -c configs/det/det_mv3_db.yml -o Global.pretrained_model="./output/det_db/best_accuracy" Global.save_inference_dir="./output/det_db_inference/"
```
The detection inference model prediction：
```shell
python3 tools/infer/predict_det.py --det_algorithm="DB" --det_model_dir="./output/det_db_inference/" --image_dir="./doc/imgs/" --use_gpu=True
```
If it is other detection algorithms, such as the EAST, the det_algorithm parameter needs to be modified to EAST, and the default is the DB algorithm:
```shell
python3 tools/infer/predict_det.py --det_algorithm="EAST" --det_model_dir="./output/det_db_inference/" --image_dir="./doc/imgs/" --use_gpu=True
```
# 2. FAQ
Q1: The prediction results of trained model and inference model are inconsistent?
**A**: Most of the problems are caused by the inconsistency of the pre-processing and post-processing parameters during the prediction of the trained model and the pre-processing and post-processing parameters during the prediction of the inference model. Taking the model trained by the det_mv3_db.yml configuration file as an example, the solution to the problem of inconsistent prediction results between the training model and the inference model is as follows:
- Check whether the [trained model preprocessing](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L116) is consistent with the prediction [preprocessing function of the inference model](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/predict_det.py#L42). When the algorithm is evaluated, the input image size will affect the accuracy. In order to be consistent with the paper, the image is resized to [736, 1280] in the training icdar15 configuration file, but there is only a set of default parameters when the inference model predicts, which will be considered To predict the speed problem, the longest side of the image is limited to 960 for resize by default. The preprocessing function of the training model preprocessing and the inference model is located in [ppocr/data/imaug/operators.py](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/ppocr/data/imaug/operators.py#L147)
- Check whether the [post-processing of the trained model](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L51) is consistent with the [post-processing parameters of the inference](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/utility.py#L50).
doc/doc_en/environment_en.md 0 → 100644
浏览文件 @ f57a6d1b
# Environment Preparation
* [1. Python Environment Setup](#1)
+ [1.1 Windows](#1.1)
+ [1.2 Mac](#1.2)
+ [1.3 Linux](#1.3)
* [2. Install PaddlePaddle 2.0](#2)
<a name="1"></a>
## 1. Python Environment Setup
<a name="1.1"></a>
### 1.1 Windows
#### 1.1.1 Install Anaconda
- Note: To use paddlepaddle you need to install python environment first, here we choose python integrated environment Anaconda toolkit
- Anaconda is a common python package manager
- After installing Anaconda, you can install the python environment, as well as numpy and other required toolkit environment.
- Anaconda download.
- Address: https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
- Most Win10 computers are 64-bit operating systems, choose x86_64 version; if the computer is a 32-bit operating system, choose x86.exe
<img src="../install/windows/Anaconda_download.png" alt="anaconda download" width="800" align="center "/>
- After the download is complete, double-click the installer to enter the graphical interface
- The default installation location is C drive, it is recommended to change the installation location to D drive.
<img src="../install/windows/anaconda_install_folder.png" alt="install config" width="500" align=" left"/>
- Check conda to add environment variables and ignore the warning that
<img src="../install/windows/anaconda_install_env.png" alt="add conda to path" width="500" align="center"/>
#### 1.1.2 Opening the terminal and creating the conda environment
- Open Anaconda Prompt terminal: bottom left Windows Start Menu -> Anaconda3 -> Anaconda Prompt start console
<img src="../install/windows/anaconda_prompt.png" alt="anaconda download" width="300" align="center"/>
- Create a new conda environment
```shell
# Enter the following command at the command line to create an environment named paddle_env
# Here to speed up the download, use the Tsinghua source
conda create --name paddle_env python=3.8 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ # This is a one line command
```
This command will create an executable environment named paddle_env with python version 3.8, which will take a while depending on the network status
The command line will then output a prompt, type y and enter to continue the installation
<img src="../install/windows/conda_new_env.png" alt="conda create" width="700" align="center"/>
- To activate the conda environment you just created, enter the following command at the command line.
```shell
# Activate the paddle_env environment
conda activate paddle_env
# View the current location of python
where python
```
<img src="../install/windows/conda_list_env.png" alt="create environment" width="600" align="center"/>
The above anaconda environment and python environment are installed
<a name="1.2"></a>
### 1.2 Mac
#### 1.2.1 Installing Anaconda
- Note: To use paddlepaddle you need to install the python environment first, here we choose the python integrated environment Anaconda toolkit
- Anaconda is a common python package manager
- After installing Anaconda, you can install the python environment, as well as numpy and other required toolkit environment
- Anaconda download:.
- Address: https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
<img src="../install/mac/anaconda_start.png" alt="anaconda download" width="800" align="center"/>
- Select `Anaconda3-2021.05-MacOSX-x86_64.pkg` at the bottom to download
- After downloading, double click on the .pkg file to enter the graphical interface
- Just follow the default settings, it will take a while to install
- It is recommended to install a code editor such as vscode or pycharm
#### 1.2.2 Open a terminal and create a conda environment
- Open the terminal
- Press command and spacebar at the same time, type "terminal" in the focus search, double click to enter terminal
- **Add conda to the environment variables**
- Environment variables are added so that the system can recognize the conda command
- Open `~/.bash_profile` in the terminal by typing the following command.
```shell
vim ~/.bash_profile
```
- Add conda as an environment variable in `~/.bash_profile`.
```shell
# Press i first to enter edit mode
# In the first line type.
export PATH="~/opt/anaconda3/bin:$PATH"
# If you customized the installation location during installation, change ~/opt/anaconda3/bin to the bin folder in the customized installation directory
```
```shell
# The modified ~/.bash_profile file should look like this (where xxx is the username)
export PATH="~/opt/anaconda3/bin:$PATH"
# >>> conda initialize >>>
# !!! Contents within this block are managed by 'conda init' !!!
__conda_setup="$('/Users/xxx/opt/anaconda3/bin/conda' 'shell.bash' 'hook' 2> /dev/null)"
if [ $? -eq 0 ]; then
eval "$__conda_setup"
else
if [ -f "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh" ]; then
. "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh"
else
export PATH="/Users/xxx/opt/anaconda3/bin:$PATH"
fi
fi
unset __conda_setup
# <<< conda initialize <<<
```
- When you are done, press `esc` to exit edit mode, then type `:wq!` and enter to save and exit
- Verify that the conda command is recognized.
- Enter `source ~/.bash_profile` in the terminal to update the environment variables
- Enter `conda info --envs` in the terminal again, if it shows that there is a base environment, then conda has been added to the environment variables
- Create a new conda environment
```shell
# Enter the following command at the command line to create an environment called paddle_env
# Here to speed up the download, use Tsinghua source
conda create --name paddle_env python=3.8 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
```
- This command will create an executable environment named paddle_env with python version 3.8, which will take a while depending on the network status
- The command line will then output a prompt, type y and enter to continue the installation
- <img src="../install/mac/conda_create.png" alt="conda_create" width="600" align="center"/>
- To activate the conda environment you just created, enter the following command at the command line.
```shell
# Activate the paddle_env environment
conda activate paddle_env
# View the current location of python
where python
```
<img src="../install/mac/conda_activate.png" alt="conda_actviate" width="600" align="center"/>
The above anaconda environment and python environment are installed
<a name="1.3"></a>
### 1.3 Linux
Linux users can choose to run either Anaconda or Docker. If you are familiar with Docker and need to train the PaddleOCR model, it is recommended to use the Docker environment, where the development process of PaddleOCR is run. If you are not familiar with Docker, you can also use Anaconda to run the project.
#### 1.3.1 Anaconda environment configuration
- Note: To use paddlepaddle you need to install the python environment first, here we choose the python integrated environment Anaconda toolkit
- Anaconda is a common python package manager
- After installing Anaconda, you can install the python environment, as well as numpy and other required toolkit environment
- **Download Anaconda**.
- Download at: https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
<img src="../install/linux/anaconda_download.png" akt="anaconda download" width="800" align="center"/>
- Select the appropriate version for your operating system
- Type `uname -m` in the terminal to check the command set used by your system
- Download method 1: Download locally, then transfer the installation package to the linux server
- Download method 2: Directly use linux command line to download
```shell
# First install wget
sudo apt-get install wget # Ubuntu
sudo yum install wget # CentOS
```
```bash
# Then use wget to download from Tsinghua source
# If you want to download Anaconda3-2021.05-Linux-x86_64.sh, the download command is as follows
wget https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/Anaconda3-2021.05-Linux-x86_64.sh
# If you want to download another version, you need to change the file name after the last 1 / to the version you want to download
```
- To install Anaconda.
- Type `sh Anaconda3-2021.05-Linux-x86_64.sh` at the command line
- If you downloaded a different version, replace the file name of the command with the name of the file you downloaded
- Just follow the installation instructions
- You can exit by typing q when viewing the license
- **Add conda to the environment variables**
- If you have already added conda to the environment variable path during the installation, you can skip this step
- Open `~/.bashrc` in a terminal.
```shell
# Enter the following command in the terminal.
vim ~/.bashrc
```
- Add conda as an environment variable in `~/.bashrc`.
```shell
# Press i first to enter edit mode # In the first line enter.
export PATH="~/anaconda3/bin:$PATH"
# If you customized the installation location during installation, change ~/anaconda3/bin to the bin folder in the customized installation directory
```
```shell
# The modified ~/.bash_profile file should look like this (where xxx is the username)
export PATH="~/opt/anaconda3/bin:$PATH"
# >>> conda initialize >>>
# !!! Contents within this block are managed by 'conda init' !!!
__conda_setup="$('/Users/xxx/opt/anaconda3/bin/conda' 'shell.bash' 'hook' 2> /dev/null)"
if [ $? -eq 0 ]; then
eval "$__conda_setup"
else
if [ -f "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh" ]; then
. "/Users/xxx/opt/anaconda3/etc/profile.d/conda.sh"
else
export PATH="/Users/xxx/opt/anaconda3/bin:$PATH"
fi
fi
unset __conda_setup
# <<< conda initialize <<<
```
- When you are done, press `esc` to exit edit mode, then type `:wq!` and enter to save and exit
- Verify that the conda command is recognized.
- Enter `source ~/.bash_profile` in the terminal to update the environment variables
- Enter `conda info --envs` in the terminal again, if it shows that there is a base environment, then conda has been added to the environment variables
- Create a new conda environment
```shell
# Enter the following command at the command line to create an environment called paddle_env
# Here to speed up the download, use Tsinghua source
conda create --name paddle_env python=3.8 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
```
- This command will create an executable environment named paddle_env with python version 3.8, which will take a while depending on the network status
- The command line will then output a prompt, type y and enter to continue the installation
<img src="../install/linux/conda_create.png" alt="conda_create" width="500" align="center"/>
- To activate the conda environment you just created, enter the following command at the command line.
```shell
# Activate the paddle_env environment
conda activate paddle_env
```
The above anaconda environment and python environment are installed
#### 1.3.2 Docker environment preparation
**The first time you use this docker image, it will be downloaded automatically. Please be patient.**
```bash
# Switch to the working directory
cd /home/Projects
# You need to create a docker container for the first run, and do not need to run the current command when you run it again
# Create a docker container named ppocr and map the current directory to the /paddle directory of the container
# If using CPU, use docker instead of nvidia-docker to create docker
sudo docker run --name ppocr -v $PWD:/paddle --network=host -it paddlepaddle/paddle:latest-dev-cuda10.1-cudnn7-gcc82 /bin/bash
```
<a name="2"></a>
## 2. Install PaddlePaddle 2.0
- If you have cuda9 or cuda10 installed on your machine, please run the following command to install
```bash
python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
```
- If you only have cpu on your machine, please run the following command to install
```bash
python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
```
For more software version requirements, please refer to the instructions in [Installation Document](https://www.paddlepaddle.org.cn/install/quick) for operation.
doc/doc_en/inference_ppocr_en.md 0 → 100755
浏览文件 @ f57a6d1b
# Reasoning based on Python prediction engine
This article introduces the use of the Python inference engine for the PP-OCR model library. The content is in order of text detection, text recognition, direction classifier and the prediction method of the three in series on the CPU and GPU.
- [TEXT DETECTION MODEL INFERENCE](#DETECTION_MODEL_INFERENCE)
- [TEXT RECOGNITION MODEL INFERENCE](#RECOGNITION_MODEL_INFERENCE)
- [1. LIGHTWEIGHT CHINESE MODEL](#LIGHTWEIGHT_RECOGNITION)
- [2. MULTILINGUAL MODEL INFERENCE](MULTILINGUAL_MODEL_INFERENCE)
- [ANGLE CLASSIFICATION MODEL INFERENCE](#ANGLE_CLASS_MODEL_INFERENCE)
- [TEXT DETECTION ANGLE CLASSIFICATION AND RECOGNITION INFERENCE CONCATENATION](#CONCATENATION)
<a name="DETECTION_MODEL_INFERENCE"></a>
## TEXT DETECTION MODEL INFERENCE
The default configuration is based on the inference setting of the DB text detection model. For lightweight Chinese detection model inference, you can execute the following commands:
```
# download DB text detection inference model
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar
tar xf ch_ppocr_mobile_v2.0_det_infer.tar
# predict
python3 tools/infer/predict_det.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/"
```
The visual text detection results are saved to the ./inference_results folder by default, and the name of the result file is prefixed with'det_res'. Examples of results are as follows:
![](../imgs_results/det_res_00018069.jpg)
You can use the parameters `limit_type` and `det_limit_side_len` to limit the size of the input image,
The optional parameters of `limit_type` are [`max`, `min`], and
`det_limit_size_len` is a positive integer, generally set to a multiple of 32, such as 960.
The default setting of the parameters is `limit_type='max', det_limit_side_len=960`. Indicates that the longest side of the network input image cannot exceed 960,
If this value is exceeded, the image will be resized with the same width ratio to ensure that the longest side is `det_limit_side_len`.
Set as `limit_type='min', det_limit_side_len=960`, it means that the shortest side of the image is limited to 960.
If the resolution of the input picture is relatively large and you want to use a larger resolution prediction, you can set det_limit_side_len to the desired value, such as 1216:
```
python3 tools/infer/predict_det.py --image_dir="./doc/imgs/1.jpg" --det_model_dir="./inference/det_db/" --det_limit_type=max --det_limit_side_len=1216
```
If you want to use the CPU for prediction, execute the command as follows
```
python3 tools/infer/predict_det.py --image_dir="./doc/imgs/1.jpg" --det_model_dir="./inference/det_db/" --use_gpu=False
```
<a name="RECOGNITION_MODEL_INFERENCE"></a>
## TEXT RECOGNITION MODEL INFERENCE
<a name="LIGHTWEIGHT_RECOGNITION"></a>
### 1. LIGHTWEIGHT CHINESE TEXT RECOGNITION MODEL REFERENCE
For lightweight Chinese recognition model inference, you can execute the following commands:
```
# download CRNN text recognition inference model
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar
tar xf ch_ppocr_mobile_v2.0_rec_infer.tar
python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_10.png" --rec_model_dir="ch_ppocr_mobile_v2.0_rec_infer"
```
![](../imgs_words_en/word_10.png)
After executing the command, the prediction results (recognized text and score) of the above image will be printed on the screen.
```bash
Predicts of ./doc/imgs_words_en/word_10.png:('PAIN', 0.9897658)
```
<a name="MULTILINGUAL_MODEL_INFERENCE"></a>
### 2. MULTILINGAUL MODEL INFERENCE
If you need to predict other language models, when using inference model prediction, you need to specify the dictionary path used by `--rec_char_dict_path`. At the same time, in order to get the correct visualization results,
You need to specify the visual font path through `--vis_font_path`. There are small language fonts provided by default under the `doc/fonts` path, such as Korean recognition:
```
python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words/korean/1.jpg" --rec_model_dir="./your inference model" --rec_char_type="korean" --rec_char_dict_path="ppocr/utils/dict/korean_dict.txt" --vis_font_path="doc/fonts/korean.ttf"
```
![](../imgs_words/korean/1.jpg)
After executing the command, the prediction result of the above figure is:
``` text
Predicts of ./doc/imgs_words/korean/1.jpg:('바탕으로', 0.9948904)
```
<a name="ANGLE_CLASS_MODEL_INFERENCE"></a>
## ANGLE CLASSIFICATION MODEL INFERENCE
For angle classification model inference, you can execute the following commands:
```
# download text angle class inference model：
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar
tar xf ch_ppocr_mobile_v2.0_cls_infer.tar
python3 tools/infer/predict_cls.py --image_dir="./doc/imgs_words_en/word_10.png" --cls_model_dir="ch_ppocr_mobile_v2.0_cls_infer"
```
![](../imgs_words_en/word_10.png)
After executing the command, the prediction results (classification angle and score) of the above image will be printed on the screen.
```
Predicts of ./doc/imgs_words_en/word_10.png:['0', 0.9999995]
```
<a name="CONCATENATION"></a>
## TEXT DETECTION ANGLE CLASSIFICATION AND RECOGNITION INFERENCE CONCATENATION
When performing prediction, you need to specify the path of a single image or a folder of images through the parameter `image_dir`, the parameter `det_model_dir` specifies the path to detect the inference model, the parameter `cls_model_dir` specifies the path to angle classification inference model and the parameter `rec_model_dir` specifies the path to identify the inference model. The parameter `use_angle_cls` is used to control whether to enable the angle classification model. The parameter `use_mp` specifies whether to use multi-process to infer `total_process_num` specifies process number when using multi-process. The parameter . The visualized recognition results are saved to the `./inference_results` folder by default.
```shell
# use direction classifier
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/" --cls_model_dir="./inference/cls/" --rec_model_dir="./inference/rec_crnn/" --use_angle_cls=true
# not use use direction classifier
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/"
# use multi-process
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/00018069.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/" --use_angle_cls=false --use_mp=True --total_process_num=6
```
After executing the command, the recognition result image is as follows:
![](../imgs_results/system_res_00018069.jpg)
doc/doc_en/models_and_config_en.md 0 → 100644
浏览文件 @ f57a6d1b
# PP-OCR Model and Configuration
- [INTRODUCTION ABOUT OCR](#introduction-about-ocr)
* [BASIC CONCEPTS OF OCR DETECTION MODEL](#basic-concepts-of-ocr-detection-model)
* [Basic concepts of OCR recognition model](#basic-concepts-of-ocr-recognition-model)
* [PP-OCR model](#pp-ocr-model)
* [And a table of contents](#and-a-table-of-contents)
* [On the right](#on-the-right)
## 1. INTRODUCTION ABOUT OCR
This section briefly introduces the basic concepts of OCR detection model and recognition model, and introduces PaddleOCR's PP-OCR model.
OCR (Optical Character Recognition, Optical Character Recognition) is currently the general term for text recognition. It is not limited to document or book text recognition, but also includes recognizing text in natural scenes. It can also be called STR (Scene Text Recognition).
OCR text recognition generally includes two parts, text detection and text recognition. The text detection module first uses detection algorithms to detect text lines in the image. And then the recognition algorithm to identify the specific text in the text line.
### 1.1 BASIC CONCEPTS OF OCR DETECTION MODEL
Text detection can locate the text area in the image, and then usually mark the word or text line in the form of a bounding box. Traditional text detection algorithms mostly extract features manually, which are characterized by fast speed and good effect in simple scenes, but the effect will be greatly reduced when faced with natural scenes. Currently, deep learning methods are mostly used.
Text detection algorithms based on deep learning can be roughly divided into the following categories:
1. Method based on target detection. Generally, after the text box is predicted, the final text box is filtered through NMS, which is mostly four-point text box, which is not ideal for curved text scenes. Typical algorithms are methods such as EAST and Text Box.
2. Method based on text segmentation. The text line is regarded as the segmentation target, and then the external text box is constructed through the segmentation result, which can handle curved text, and the effect is not ideal for the text cross scene problem. Typical algorithms are DB, PSENet and other methods.
3. Hybrid target detection and segmentation method.
### 1.2 Basic concepts of OCR recognition model
The input of the OCR recognition algorithm is generally text lines images which has less background information, and the text information occupies the main part. The recognition algorithm can be divided into two types of algorithms:
1. CTC-based method. The text prediction module of the recognition algorithm is based on CTC, and the commonly used algorithm combination is CNN+RNN+CTC. There are also some algorithms that try to add transformer modules to the network and so on.
2. Attention-based method. The text prediction module of the recognition algorithm is based on Attention, and the commonly used algorithm combination is CNN+RNN+Attention.
### 1.3 PP-OCR model
PaddleOCR integrates many OCR algorithms, text detection algorithms include DB, EAST, SAST, etc., text recognition algorithms include CRNN, RARE, StarNet, Rosetta, SRN and other algorithms.
Among them, PaddleOCR has released the PP-OCR series model for the general OCR in Chinese and English natural scenes. The PP-OCR model is composed of the DB+CRNN algorithm. It uses massive Chinese data training and model tuning methods to have high text detection and recognition capabilities in Chinese scenes. And PaddleOCR has launched a high-precision and ultra-lightweight PP-OCRv2 model. The detection model is only 3M, and the recognition model is only 8.5M. Using [PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim)'s model quantification method, the detection model can be compressed to 0.8M without reducing the accuracy. The recognition is compressed to 3M, which is more suitable for mobile deployment scenarios.
doc/doc_en/models_list_en.md
浏览文件 @ f57a6d1b
## OCR model list（V2.0, updated on 2021.1.20）
## OCR model list（V2.1, updated on 2021.9.6）
> **Note**
> 1. Compared with [models 1.1](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/models_list_en.md), which are trained with static graph programming paradigm, models 2.0 are the dynamic graph trained version and achieve close performance.
> 2. All models in this tutorial are all ppocr-series models, for more introduction of algorithms and models based on public dataset, you can refer to [algorithm overview tutorial](./algorithm_overview_en.md).
> 1. Compared with the model v2.0, the 2.1 version of the detection model has a improvement in accuracy, and the 2.1 version of the recognition model is optimized in accuracy and CPU speed.
> 2. Compared with [models 1.1](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/models_list_en.md), which are trained with static graph programming paradigm, models 2.0 are the dynamic graph trained version and achieve close performance.
> 3. All models in this tutorial are all ppocr-series models, for more introduction of algorithms and models based on public dataset, you can refer to [algorithm overview tutorial](./algorithm_overview_en.md).
- [1. Text Detection Model](#Detection)
- [2. Text Recognition Model](#Recognition)
......@@ -28,6 +29,8 @@ Relationship of the above models is as follows.
|model name|description|config|model size|download|
| --- | --- | --- | --- | --- |
|ch_ppocr_mobile_slim_v2.1_det|slim quantization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_lite_train_cml_v2.1.yml](../../configs/det/ch_ppocr_v2.1/ch_det_lite_train_cml_v2.1.yml)| 3M |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_slim_quant_infer.tar)|
|ch_ppocr_mobile_v2.1_det|Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_lite_train_cml_v2.1.ym](../../configs/det/ch_ppocr_v2.1/ch_det_lite_train_cml_v2.1.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_distill_train.tar)|
|ch_ppocr_mobile_slim_v2.0_det|Slim pruned lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|2.6M |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_prune_infer.tar)|
|ch_ppocr_mobile_v2.0_det|Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|
|ch_ppocr_server_v2.0_det|General model, which is larger than the lightweight model, but achieved better performance|[ch_det_res18_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_res18_db_v2.0.yml)|47M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)|
......@@ -40,6 +43,8 @@ Relationship of the above models is as follows.
|model name|description|config|model size|download|
| --- | --- | --- | --- | --- |
|ch_ppocr_mobile_slim_v2.1_rec|Slim qunatization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[rec_chinese_lite_train_distillation_v2.1.yml](../../configs/rec/ch_ppocr_v2.1/rec_chinese_lite_train_distillation_v2.1.yml)| 9M |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_slim_quant_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_slim_quant_train.tar) |
|ch_ppocr_mobile_v2.1_rec|Original lightweight model, supporting Chinese, English, multilingual text detection|[rec_chinese_lite_train_distillation_v2.1.yml](../../configs/rec/ch_ppocr_v2.1/rec_chinese_lite_train_distillation_v2.1.yml)|8.5M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_train.tar) |
|ch_ppocr_mobile_slim_v2.0_rec|Slim pruned and quantized lightweight model, supporting Chinese, English and number recognition|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)| 6M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_train.tar) |
|ch_ppocr_mobile_v2.0_rec|Original lightweight model, supporting Chinese, English and number recognition|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)|5.2M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
|ch_ppocr_server_v2.0_rec|General model, supporting Chinese, English and number recognition|[rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml)|94.8M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
......@@ -58,45 +63,6 @@ Relationship of the above models is as follows.
<a name="Multilingual"></a>
#### Multilingual Recognition Model（Updating...）
**Note：** The configuration file of the new multi language model is generated by code. You can use the `--help` parameter to check which multi language are supported by current PaddleOCR.
```bash
# The code needs to run in the specified directory
cd {your/path/}PaddleOCR/configs/rec/multi_language/
python3 generate_multi_language_configs.py --help
```
Take the Italian configuration file as an example：
##### 1.Generate Italian configuration file to test the model provided
you can generate the default configuration file through the following command, and use the default language dictionary provided by paddleocr for prediction.
```bash
# The code needs to run in the specified directory
cd {your/path/}PaddleOCR/configs/rec/multi_language/
# Set the required language configuration file through -l or --language parameter
# This command will write the default parameter to the configuration file.
python3 generate_multi_language_configs.py -l it
```
##### 2. Generate Italian configuration file to train your own data
If you want to train your own model, you can prepare the training set file, verification set file, dictionary file and training data path. Here we assume that the Italian training set, verification set, dictionary and training data path are:
- Training set:{your/path/}PaddleOCR/train_data/train_list.txt
- Validation set: {your/path/}PaddleOCR/train_data/val_list.txt
- Use the default dictionary provided by paddleocr:{your/path/}PaddleOCR/ppocr/utils/dict/it_dict.txt
- Training data path:{your/path/}PaddleOCR/train_data
```bash
# The code needs to run in the specified directory
cd {your/path/}PaddleOCR/configs/rec/multi_language/
# The -l or --language parameter is required
# --train modify train_list path
# --val modify eval_list path
# --data_dir modify data dir
# -o modify default parameters
# --dict Change the dictionary path. The example uses the default dictionary path, so that this parameter can be empty.
python3 generate_multi_language_configs.py -l it \
--train {path/to/train_list} \
--val {path/to/val_list} \
--data_dir {path/to/data_dir} \
-o Global.use_gpu=False
```
|model name| dict file | description|config|model size|download|
| --- | --- | --- |--- | --- | --- |
| french_mobile_v2.0_rec | ppocr/utils/dict/french_dict.txt | Lightweight model for French recognition|[rec_french_lite_train.yml](../../configs/rec/multi_language/rec_french_lite_train.yml)|2.65M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/french_mobile_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/french_mobile_v2.0_rec_train.tar) |
......@@ -120,12 +86,14 @@ For more supported languages, please refer to : [Multi-language model](./multi_l
|model name|description|config|model size|download|
| --- | --- | --- | --- | --- |
|ch_ppocr_mobile_slim_v2.0_cls|Slim quantized model|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_train.tar) |
|ch_ppocr_mobile_v2.0_cls|Original model|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)|1.38M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |
|ch_ppocr_mobile_slim_v2.0_cls|Slim quantized model for text angle classification|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_train.tar) |
|ch_ppocr_mobile_v2.0_cls|Original model for text angle classification|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)|1.38M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |
<a name="Paddle-Lite"></a>
### 4. Paddle-Lite Model
|Version|Introduction|Model size|Detection model|Text Direction model|Recognition model|Paddle-Lite branch|
|---|---|---|---|---|---|---|
|V2.0|extra-lightweight chinese OCR optimized model|7.8M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
|V2.0(slim)|extra-lightweight chinese OCR optimized model|3.3M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_slim_opt.nb)|v2.9|
|V2.1|ppocr_v2.1 extra-lightweight chinese OCR optimized model|11M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_infer_opt.nb)|v2.9|
|V2.1(slim)|extra-lightweight chinese OCR optimized model|4.9M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_slim_opt.nb)|v2.9|
|V2.0|ppocr_v2.0 extra-lightweight chinese OCR optimized model|7.8M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
|V2.0(slim)|ppovr_v2.0 extra-lightweight chinese OCR optimized model|3.3M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_slim_opt.nb)|v2.9|
doc/doc_en/multi_languages_en.md
浏览文件 @ f57a6d1b
......@@ -198,13 +198,13 @@ If necessary, you can read related documents:
| Language | Abbreviation | | Language | Abbreviation |
| --- | --- | --- | --- | --- |
|chinese and english|ch| |Arabic|ar|
|english|en| |Hindi|hi|
|french|fr| |Uyghur|ug|
|german|german| |Persian|fa|
|japan|japan| |Urdu|ur|
|korean|korean| | Serbian(latin) |rs_latin|
|chinese traditional |ch_tra| |Occitan |oc|
|Chinese & English|ch| |Arabic|ar|
|English|en| |Hindi|hi|
|French|fr| |Uyghur|ug|
|German|german| |Persian|fa|
|Japan|japan| |Urdu|ur|
|Korean|korean| | Serbian(latin) |rs_latin|
|Chinese Traditional |chinese_cht| |Occitan |oc|
| Italian |it| |Marathi|mr|
|Spanish |es| |Nepali|ne|
| Portuguese|pt| |Serbian(cyrillic)|rs_cyrillic|
......
doc/doc_en/paddleOCR_overview_en.md 0 → 100644
浏览文件 @ f57a6d1b
# PaddleOCR Overview and Project Clone
## 1. PaddleOCR Overview
PaddleOCR contains rich text detection, text recognition and end-to-end algorithms. Combining actual testing and industrial experience, PaddleOCR chooses DB and CRNN as the basic detection and recognition models, and proposes a series of models, named PP-OCR, for industrial applications after a series of optimization strategies. The PP-OCR model is aimed at general scenarios and forms a model library according to different languages. Based on the capabilities of PP-OCR, PaddleOCR releases the PP-Structure tool library for document scene tasks, including two major tasks: layout analysis and table recognition. In order to get through the entire process of industrial landing, PaddleOCR provides large-scale data production tools and a variety of prediction deployment tools to help developers quickly turn ideas into reality.
<div align="center">
<img src="../overview_en.png">
</div>
## 2. Project Clone
### **2.1 Clone PaddleOCR repo**
```
# Recommend
git clone https://github.com/PaddlePaddle/PaddleOCR
# If you cannot pull successfully due to network problems, you can also choose to use the code hosting on the cloud:
git clone https://gitee.com/paddlepaddle/PaddleOCR
# Note: The cloud-hosting code may not be able to synchronize the update with this GitHub project in real time. There might be a delay of 3-5 days. Please give priority to the recommended method.
```
### **2.2 Install third-party libraries**
```
cd PaddleOCR
pip3 install -r requirements.txt
```
If you getting this error `OSError: [WinError 126] The specified module could not be found` when you install shapely on windows.
Please try to download Shapely whl file using [http://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely](http://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely).
Reference: [Solve shapely installation on windows](
\ No newline at end of file
doc/doc_en/quickstart_en.md
浏览文件 @ f57a6d1b
# Quick start of Chinese OCR model
# PaddleOCR Quick Start
## 1. Prepare for the environment
[PaddleOCR Quick Start](#paddleocr-quick-start)
Please refer to [quick installation](./installation_en.md) to configure the PaddleOCR operating environment.
+ [1. Install PaddleOCR Whl Package](#1-install-paddleocr-whl-package)
* [2. Easy-to-Use](#2-easy-to-use)
+ [2.1 Use by command line](#21-use-by-command-line)
- [2.1.1 English and Chinese Model](#211-english-and-chinese-model)
- [2.1.2 Multi-language Model](#212-multi-language-model)
- [2.1.3 Layout Analysis](#213-layoutAnalysis)
+ [2.2 Use by Code](#22-use-by-code)
- [2.2.1 Chinese & English Model and Multilingual Model](#221-chinese---english-model-and-multilingual-model)
- [2.2.2 Layout Analysis](#222-layoutAnalysis)
* Note: Support the use of PaddleOCR through whl package installation，pelease refer [PaddleOCR Package](./whl_en.md).
## 2.inference models
The detection and recognition models on the mobile and server sides are as follows. For more models (including multiple languages), please refer to [PP-OCR v2.0 series model list](../doc_ch/models_list.md)
<a name="1-install-paddleocr-whl-package"></a>
| Model introduction | Model name | Recommended scene | Detection model | Direction Classifier | Recognition model |
| ------------ | --------------- | ----------------|---- | ---------- | -------- |
| Ultra-lightweight Chinese OCR model (8.1M) | ch_ppocr_mobile_v2.0_xx |Mobile-side/Server-side|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [pretrained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pretrained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [pretrained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
| Universal Chinese OCR model (143M) | ch_ppocr_server_v2.0_xx |Server-side |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [pretrained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pretrained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [pretrained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |
## 1. Install PaddleOCR Whl Package
```bash
pip install "paddleocr>=2.0.1" # Recommend to use version 2.0.1+
```
* If `wget` is not installed in the windows environment, you can copy the link to the browser to download when downloading the model, then uncompress it and place it in the corresponding directory.
- **For windows users:** If you getting this error `OSError: [WinError 126] The specified module could not be found` when you install shapely on windows. Please try to download Shapely whl file [here](http://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely).
Copy the download address of the `inference model` for detection and recognition in the table above, and uncompress them.
Reference: [Solve shapely installation on windows](https://stackoverflow.com/questions/44398265/install-shapely-oserror-winerror-126-the-specified-module-could-not-be-found)
```
mkdir inference && cd inference
# Download the detection model and unzip
wget {url/of/detection/inference_model} && tar xf {name/of/detection/inference_model/package}
# Download the recognition model and unzip
wget {url/of/recognition/inference_model} && tar xf {name/of/recognition/inference_model/package}
# Download the direction classifier model and unzip
wget {url/of/classification/inference_model} && tar xf {name/of/classification/inference_model/package}
cd ..
```
- **For layout analysis users**, run the following command to install **Layout-Parser**
Take the ultra-lightweight model as an example:
```bash
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
```
<a name="2-easy-to-use"></a>
## 2. Easy-to-Use
<a name="21-use-by-command-line"></a>
### 2.1 Use by command line
PaddleOCR provides a series of test images, click [here](https://paddleocr.bj.bcebos.com/dygraph_v2.1/ppocr_img.zip) to download, and then switch to the corresponding directory in the terminal
```bash
cd /path/to/ppocr_img
```
mkdir inference && cd inference
# Download the detection model of the ultra-lightweight Chinese OCR model and uncompress it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_det_infer.tar
# Download the recognition model of the ultra-lightweight Chinese OCR model and uncompress it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar
# Download the angle classifier model of the ultra-lightweight Chinese OCR model and uncompress it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar && tar xf ch_ppocr_mobile_v2.0_cls_infer.tar
cd ..
```
After decompression, the file structure should be as follows:
If you do not use the provided test image, you can replace the following `--image_dir` parameter with the corresponding test image path
<a name="211-english-and-chinese-model"></a>
#### 2.1.1 Chinese and English Model
* Detection, direction classification and recognition: set the direction classifier parameter`--use_angle_cls true` to recognize vertical text.
```bash
paddleocr --image_dir ./imgs_en/img_12.jpg --use_angle_cls true --lang en
```
Output will be a list, each item contains bounding box, text and recognition confidence
```bash
[[[442.0, 173.0], [1169.0, 173.0], [1169.0, 225.0], [442.0, 225.0]], ['ACKNOWLEDGEMENTS', 0.99283075]]
[[[393.0, 340.0], [1207.0, 342.0], [1207.0, 389.0], [393.0, 387.0]], ['We would like to thank all the designers and', 0.9357758]]
[[[399.0, 398.0], [1204.0, 398.0], [1204.0, 433.0], [399.0, 433.0]], ['contributors whohave been involved in the', 0.9592447]]
......
```
* Only detection: set `--rec` to `false`
```bash
paddleocr --image_dir ./imgs_en/img_12.jpg --rec false
```
Output will be a list, each item only contains bounding box
```bash
[[756.0, 812.0], [805.0, 812.0], [805.0, 830.0], [756.0, 830.0]]
[[820.0, 803.0], [1085.0, 801.0], [1085.0, 836.0], [820.0, 838.0]]
[[393.0, 801.0], [715.0, 805.0], [715.0, 839.0], [393.0, 836.0]]
......
```
* Only recognition: set `--det` to `false`
```bash
paddleocr --image_dir ./imgs_words_en/word_10.png --det false --lang en
```
Output will be a list, each item contains text and recognition confidence
```bash
['PAIN', 0.990372]
```
More whl package usage can be found in [whl package](./whl_en.md)
<a name="212-multi-language-model"></a>
#### 2.1.2 Multi-language Model
Paddleocr currently supports 80 languages, which can be switched by modifying the `--lang` parameter.
``` bash
paddleocr --image_dir ./doc/imgs_en/254.jpg --lang=en
```
├── ch_ppocr_mobile_v2.0_cls_infer
│ ├── inference.pdiparams
│ ├── inference.pdiparams.info
│ └── inference.pdmodel
├── ch_ppocr_mobile_v2.0_det_infer
│ ├── inference.pdiparams
│ ├── inference.pdiparams.info
│ └── inference.pdmodel
├── ch_ppocr_mobile_v2.0_rec_infer
├── inference.pdiparams
├── inference.pdiparams.info
└── inference.pdmodel
<div align="center">
<img src="../imgs_en/254.jpg" width="300" height="600">
<img src="../imgs_results/multi_lang/img_02.jpg" width="600" height="600">
</div>
The result is a list, each item contains a text box, text and recognition confidence
```text
[('PHO CAPITAL', 0.95723116), [[66.0, 50.0], [327.0, 44.0], [327.0, 76.0], [67.0, 82.0]]]
[('107 State Street', 0.96311164), [[72.0, 90.0], [451.0, 84.0], [452.0, 116.0], [73.0, 121.0]]]
[('Montpelier Vermont', 0.97389287), [[69.0, 132.0], [501.0, 126.0], [501.0, 158.0], [70.0, 164.0]]]
[('8022256183', 0.99810505), [[71.0, 175.0], [363.0, 170.0], [364.0, 202.0], [72.0, 207.0]]]
[('REG 07-24-201706:59 PM', 0.93537045), [[73.0, 299.0], [653.0, 281.0], [654.0, 318.0], [74.0, 336.0]]]
[('045555', 0.99346405), [[509.0, 331.0], [651.0, 325.0], [652.0, 356.0], [511.0, 362.0]]]
[('CT1', 0.9988654), [[535.0, 367.0], [654.0, 367.0], [654.0, 406.0], [535.0, 406.0]]]
......
```
## 3. Single image or image set prediction
Commonly used multilingual abbreviations include
| Language | Abbreviation | | Language | Abbreviation | | Language | Abbreviation |
| ------------------- | ------------ | ---- | -------- | ------------ | ---- | -------- | ------------ |
| Chinese & English | ch | | French | fr | | Japanese | japan |
| English | en | | German | german | | Korean | korean |
| Chinese Traditional | chinese_cht | | Italian | it | | Russian | ru |
* The following code implements text detection、angle class and recognition process. When performing prediction, you need to specify the path of a single image or image set through the parameter `image_dir`, the parameter `det_model_dir` specifies the path to detect the inference model, the parameter `rec_model_dir` specifies the path to identify the inference model, the parameter `use_angle_cls` specifies whether to use the direction classifier, the parameter `cls_model_dir` specifies the path to identify the direction classifier model, the parameter `use_space_char` specifies whether to predict the space char. The visual results are saved to the `./inference_results` folder by default.
A list of all languages and their corresponding abbreviations can be found in [Multi-Language Model Tutorial](./multi_languages_en.md)
<a name="213-layoutAnalysis"></a>
#### 2.1.3 Layout Analysis
Layout analysis refers to the division of 5 types of areas of the document, including text, title, list, picture and table. For the first three types of regions, directly use the OCR model to complete the text detection and recognition of the corresponding regions, and save the results in txt. For the table area, after the table structuring process, the table picture is converted into an Excel file of the same table style. The picture area will be individually cropped into an image.
To use the layout analysis function of PaddleOCR, you need to specify `--type=structure`
```bash
paddleocr --image_dir=../doc/table/1.png --type=structure
```
# Predict a single image specified by image_dir
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True
- **Results Format**
The returned results of PP-Structure is a list composed of a dict, an example is as follows
```shell
[
{ 'type': 'Text',
'bbox': [34, 432, 345, 462],
'res': ([[36.0, 437.0, 341.0, 437.0, 341.0, 446.0, 36.0, 447.0], [41.0, 454.0, 125.0, 453.0, 125.0, 459.0, 41.0, 460.0]],
[('Tigure-6. The performance of CNN and IPT models using difforen', 0.90060663), ('Tent ', 0.465441)])
}
]
```
The description of each field in dict is as follows
| Parameter | Description |
| --------- | ------------------------------------------------------------ |
| type | Type of image area |
| bbox | The coordinates of the image area in the original image, respectively [left upper x, left upper y, right bottom x, right bottom y] |
| res | OCR or table recognition result of image area。<br> Table: HTML string of the table; <br> OCR: A tuple containing the detection coordinates and recognition results of each single line of text |
- **Parameter Description：**
| Parameter | Description | Default value |
| --------------- | ------------------------------------------------------------ | -------------------------------------------- |
| output | The path where excel and recognition results are saved | ./output/table |
| table_max_len | The long side of the image is resized in table structure model | 488 |
| table_model_dir | inference model path of table structure model | None |
| table_char_type | dict path of table structure model | ../ppocr/utils/dict/table_structure_dict.txt |
<a name="22-use-by-code"></a>
### 2.2 Use by Code
<a name="221-chinese---english-model-and-multilingual-model"></a>
#### 2.2.1 Chinese & English Model and Multilingual Model
* detection, angle classification and recognition:
```python
from paddleocr import PaddleOCR,draw_ocr
# Paddleocr supports Chinese, English, French, German, Korean and Japanese.
# You can set the parameter `lang` as `ch`, `en`, `fr`, `german`, `korean`, `japan`
# to switch the language model in order.
ocr = PaddleOCR(use_angle_cls=True, lang='en') # need to run only once to download and load model into memory
img_path = './imgs_en/img_12.jpg'
result = ocr.ocr(img_path, cls=True)
for line in result:
print(line)
# draw result
from PIL import Image
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='./fonts/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
# Predict imageset specified by image_dir
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True
Output will be a list, each item contains bounding box, text and recognition confidence
# If you want to use the CPU for prediction, you need to set the use_gpu parameter to False
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True --use_gpu=False
```bash
[[[442.0, 173.0], [1169.0, 173.0], [1169.0, 225.0], [442.0, 225.0]], ['ACKNOWLEDGEMENTS', 0.99283075]]
[[[393.0, 340.0], [1207.0, 342.0], [1207.0, 389.0], [393.0, 387.0]], ['We would like to thank all the designers and', 0.9357758]]
[[[399.0, 398.0], [1204.0, 398.0], [1204.0, 433.0], [399.0, 433.0]], ['contributors whohave been involved in the', 0.9592447]]
......
```
- Universal Chinese OCR model
Visualization of results
Please follow the above steps to download the corresponding models and update the relevant parameters, The example is as follows.
<div align="center">
<img src="../imgs_results/whl/12_det_rec.jpg" width="800">
</div>
<a name="222-layoutAnalysis"></a>
```
# Predict a single image specified by image_dir
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_server_v2.0_det_infer/" --rec_model_dir="./inference/ch_ppocr_server_v2.0_rec_infer/" --cls_model_dir="./inference/ch_ppocr_mobile_v2.0_cls_infer/" --use_angle_cls=True --use_space_char=True
```
#### 2.2.2 Layout Analysis
```python
import os
import cv2
from paddleocr import PPStructure,draw_structure_result,save_structure_res
* Note
- If you want to use the recognition model which does not support space char recognition, please update the source code to the latest version and add parameters `--use_space_char=False`.
- If you do not want to use direction classifier, please update the source code to the latest version and add parameters `--use_angle_cls=False`.
table_engine = PPStructure(show_log=True)
save_folder = './output/table'
img_path = './table/1.png'
img = cv2.imread(img_path)
result = table_engine(img)
save_structure_res(result, save_folder,os.path.basename(img_path).split('.')[0])
For more text detection and recognition tandem reasoning, please refer to the document tutorial
: [Inference with Python inference engine](./inference_en.md)。
for line in result:
line.pop('img')
print(line)
In addition, the tutorial also provides other deployment methods for the Chinese OCR model:
- [Server-side C++ inference](../../deploy/cpp_infer/readme_en.md)
- [Service deployment](../../deploy/hubserving)
- [End-to-end deployment](https://github.com/PaddlePaddle/PaddleOCR/tree/develop/deploy/lite)
from PIL import Image
font_path = './fonts/simfang.ttf'
image = Image.open(img_path).convert('RGB')
im_show = draw_structure_result(image, result,font_path=font_path)
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
doc/doc_en/recognition_en.md
浏览文件 @ f57a6d1b
## TEXT RECOGNITION
# TEXT RECOGNITION
- [1 DATA PREPARATION](#DATA_PREPARATION)
- [1.1 Costom Dataset](#Costom_Dataset)
......@@ -8,8 +8,8 @@
- [2 TRAINING](#TRAINING)
- [2.1 Data Augmentation](#Data_Augmentation)
- [2.2 Training](#Training)
- [2.3 Multi-language](#Multi_language)
- [2.2 General Training](#Training)
- [2.3 Multi-language Training](#Multi_language)
- [3 EVALUATION](#EVALUATION)
......@@ -17,12 +17,12 @@
- [4.1 Training engine prediction](#Training_engine_prediction)
<a name="DATA_PREPARATION"></a>
### DATA PREPARATION
## 1 DATA PREPARATION
PaddleOCR supports two data formats:
- `LMDB` is used to train data sets stored in lmdb format;
- `general data` is used to train data sets stored in text files:
- `LMDB` is used to train data sets stored in lmdb format（LMDBDataSet）;
- `general data` is used to train data sets stored in text files（SimpleDataSet）:
Please organize the dataset as follows:
......@@ -36,7 +36,7 @@ mklink /d <path/to/paddle_ocr>/train_data/dataset <path/to/dataset>
```
<a name="Costom_Dataset"></a>
#### 1.1 Costom dataset
### 1.1 Costom dataset
If you want to use your own data for training, please refer to the following to organize your data.
......@@ -84,11 +84,12 @@ Similar to the training set, the test set also needs to be provided a folder con
```
<a name="Dataset_download"></a>
#### 1.2 Dataset download
### 1.2 Dataset download
If you do not have a dataset locally, you can download it on the official website [icdar2015](http://rrc.cvc.uab.es/?ch=4&com=downloads). Also refer to [DTRB](https://github.com/clovaai/deep-text-recognition-benchmark#download-lmdb-dataset-for-traininig-and-evaluation-from-here) ，download the lmdb format dataset required for benchmark
- ICDAR2015
If you want to reproduce the paper indicators of SRN, you need to download offline [augmented data](https://pan.baidu.com/s/1-HSZ-ZVdqBF2HaBZ5pRAKA), extraction code: y3ry. The augmented data is obtained by rotation and perturbation of mjsynth and synthtext. Please unzip the data to {your_path}/PaddleOCR/train_data/data_lmdb_Release/training/path.
If you do not have a dataset locally, you can download it on the official website [icdar2015](http://rrc.cvc.uab.es/?ch=4&com=downloads).
Also refer to [DTRB](https://github.com/clovaai/deep-text-recognition-benchmark#download-lmdb-dataset-for-traininig-and-evaluation-from-here) ，download the lmdb format dataset required for benchmark
If you want to reproduce the paper SAR, you need to download extra dataset [SynthAdd](https://pan.baidu.com/share/init?surl=uV0LtoNmcxbO-0YA7Ch4dg), extraction code: 627x. Besides, icdar2013, icdar2015, cocotext, IIIT5k datasets are also used to train. For specific details, please refer to the paper SAR.
......@@ -101,8 +102,28 @@ wget -P ./train_data/ic15_data https://paddleocr.bj.bcebos.com/dataset/rec_gt_t
wget -P ./train_data/ic15_data https://paddleocr.bj.bcebos.com/dataset/rec_gt_test.txt
```
PaddleOCR also provides a data format conversion script, which can convert ICDAR official website label to a data format
supported by PaddleOCR. The data conversion tool is in `ppocr/utils/gen_label.py`, here is the training set as an example:
```
# convert the official gt to rec_gt_label.txt
python gen_label.py --mode="rec" --input_path="{path/of/origin/label}" --output_label="rec_gt_label.txt"
```
The data format is as follows, (a) is the original picture, (b) is the Ground Truth text file corresponding to each picture:
![](../datasets/icdar_rec.png)
- Multilingual dataset
The multi-language model training method is the same as the Chinese model. The training data set is 100w synthetic data. A small amount of fonts and test data can be downloaded using the following two methods.
* [Baidu Netdisk](https://pan.baidu.com/s/1bS_u207Rm7YbY33wOECKDA) ,Extraction code:frgi.
* [Google drive](https://drive.google.com/file/d/18cSWX7wXSy4G0tbKJ0d9PuIaiwRLHpjA/view)
<a name="Dictionary"></a>
#### 1.3 Dictionary
### 1.3 Dictionary
Finally, a dictionary ({word_dict_name}.txt) needs to be provided so that when the model is trained, all the characters that appear can be mapped to the dictionary index.
......@@ -147,14 +168,26 @@ To customize the dict file, please modify the `character_dict_path` field in `co
If you need to customize dic file, please add character_dict_path field in configs/rec/rec_icdar15_train.yml to point to your dictionary path. And set character_type to ch.
<a name="Add_space_category"></a>
#### 1.4 Add space category
### 1.4 Add space category
If you want to support the recognition of the `space` category, please set the `use_space_char` field in the yml file to `True`.
**Note: use_space_char only takes effect when character_type=ch**
<a name="TRAINING"></a>
### 2 TRAINING
## 2 TRAINING
<a name="Data_Augmentation"></a>
### 2.1 Data Augmentation
PaddleOCR provides a variety of data augmentation methods. All the augmentation methods are enabled by default.
The default perturbation methods are: cvtColor, blur, jitter, Gasuss noise, random crop, perspective, color reverse, TIA augmentation.
Each disturbance method is selected with a 40% probability during the training process. For specific code implementation, please refer to: [rec_img_aug.py](../../ppocr/data/imaug/rec_img_aug.py)
<a name="Training"></a>
### 2.2 General Training
PaddleOCR provides training scripts, evaluation scripts, and prediction scripts. In this section, the CRNN recognition model will be used as an example:
......@@ -172,21 +205,15 @@ tar -xf rec_mv3_none_bilstm_ctc_v2.0_train.tar && rm -rf rec_mv3_none_bilstm_ctc
Start training:
```
# GPU training Support single card and multi-card training, specify the card number through --gpus
# GPU training Support single card and multi-card training
# Training icdar15 English data and The training log will be automatically saved as train.log under "{save_model_dir}"
#specify the single card training(Long training time, not recommended)
python3 tools/train.py -c configs/rec/rec_icdar15_train.yml
#specify the card number through --gpus
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/rec/rec_icdar15_train.yml
```
<a name="Data_Augmentation"></a>
#### 2.1 Data Augmentation
PaddleOCR provides a variety of data augmentation methods. All the augmentation methods are enabled by default.
The default perturbation methods are: cvtColor, blur, jitter, Gasuss noise, random crop, perspective, color reverse, TIA augmentation.
Each disturbance method is selected with a 40% probability during the training process. For specific code implementation, please refer to: [rec_img_aug.py](../../ppocr/data/imaug/rec_img_aug.py)
<a name="Training"></a>
#### 2.2 Training
PaddleOCR supports alternating training and evaluation. You can modify `eval_batch_step` in `configs/rec/rec_icdar15_train.yml` to set the evaluation frequency. By default, it is evaluated every 500 iter and the best acc model is saved under `output/rec_CRNN/best_accuracy` during the evaluation process.
......@@ -281,87 +308,7 @@ Eval:
**Note that the configuration file for prediction/evaluation must be consistent with the training.**
<a name="Multi_language"></a>
#### 2.3 Multi-language
PaddleOCR currently supports 80 (except Chinese) language recognition. A multi-language configuration file template is
provided under the path `configs/rec/multi_languages`: [rec_multi_language_lite_train.yml](../../configs/rec/multi_language/rec_multi_language_lite_train.yml)。
There are two ways to create the required configuration file:：
1. Automatically generated by script
[generate_multi_language_configs.py](../../configs/rec/multi_language/generate_multi_language_configs.py) Can help you generate configuration files for multi-language models
- Take Italian as an example, if your data is prepared in the following format:
```
|-train_data
|- it_train.txt # train_set label
|- it_val.txt # val_set label
|- data
|- word_001.jpg
|- word_002.jpg
|- word_003.jpg
| ...
```
You can use the default parameters to generate a configuration file:
```bash
# The code needs to be run in the specified directory
cd PaddleOCR/configs/rec/multi_language/
# Set the configuration file of the language to be generated through the -l or --language parameter.
# This command will write the default parameters into the configuration file
python3 generate_multi_language_configs.py -l it
```
- If your data is placed in another location, or you want to use your own dictionary, you can generate the configuration file by specifying the relevant parameters:
```bash
# -l or --language field is required
# --train to modify the training set
# --val to modify the validation set
# --data_dir to modify the data set directory
# --dict to modify the dict path
# -o to modify the corresponding default parameters
cd PaddleOCR/configs/rec/multi_language/
python3 generate_multi_language_configs.py -l it \ # language
--train {path/of/train_label.txt} \ # path of train_label
--val {path/of/val_label.txt} \ # path of val_label
--data_dir {train_data/path} \ # root directory of training data
--dict {path/of/dict} \ # path of dict
-o Global.use_gpu=False # whether to use gpu
...
```
Italian is made up of Latin letters, so after executing the command, you will get the rec_latin_lite_train.yml.
2. Manually modify the configuration file
You can also manually modify the following fields in the template:
```
Global:
use_gpu: True
epoch_num: 500
...
character_type: it # language
character_dict_path: {path/of/dict} # path of dict
Train:
dataset:
name: SimpleDataSet
data_dir: train_data/ # root directory of training data
label_file_list: ["./train_data/train_list.txt"] # train label path
...
Eval:
dataset:
name: SimpleDataSet
data_dir: train_data/ # root directory of val data
label_file_list: ["./train_data/val_list.txt"] # val label path
...
```
### 2.3 Multi-language Training
Currently, the multi-language algorithms supported by PaddleOCR are:
......@@ -380,9 +327,6 @@ Currently, the multi-language algorithms supported by PaddleOCR are:
For more supported languages, please refer to : [Multi-language model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md#4-support-languages-and-abbreviations)
The multi-language model training method is the same as the Chinese model. The training data set is 100w synthetic data. A small amount of fonts and test data can be downloaded using the following two methods.
* [Baidu Netdisk](https://pan.baidu.com/s/1bS_u207Rm7YbY33wOECKDA),Extraction code:frgi.
* [Google drive](https://drive.google.com/file/d/18cSWX7wXSy4G0tbKJ0d9PuIaiwRLHpjA/view)
If you want to finetune on the basis of the existing model effect, please refer to the following instructions to modify the configuration file:
......@@ -421,7 +365,7 @@ Eval:
```
<a name="EVALUATION"></a>
### 3 EVALUATION
## 3 EVALUATION
The evaluation dataset can be set by modifying the `Eval.dataset.label_file_list` field in the `configs/rec/rec_icdar15_train.yml` file.
......@@ -431,20 +375,39 @@ python3 -m paddle.distributed.launch --gpus '0' tools/eval.py -c configs/rec/rec
```
<a name="PREDICTION"></a>
### 4 PREDICTION
## 4 PREDICTION
<a name="Training_engine_prediction"></a>
#### 4.1 Training engine prediction
Using the model trained by paddleocr, you can quickly get prediction through the following script.
The default prediction picture is stored in `infer_img`, and the weight is specified via `-o Global.checkpoints`:
The default prediction picture is stored in `infer_img`, and the trained weight is specified via `-o Global.checkpoints`:
According to the `save_model_dir` and `save_epoch_step` fields set in the configuration file, the following parameters will be saved:
```
output/rec/
├── best_accuracy.pdopt
├── best_accuracy.pdparams
├── best_accuracy.states
├── config.yml
├── iter_epoch_3.pdopt
├── iter_epoch_3.pdparams
├── iter_epoch_3.states
├── latest.pdopt
├── latest.pdparams
├── latest.states
└── train.log
```
Among them, best_accuracy.* is the best model on the evaluation set; iter_epoch_x.* is the model saved at intervals of `save_epoch_step`; latest.* is the model of the last epoch.
```
# Predict English results
python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml -o Global.pretrained_model={path/to/weights}/best_accuracy Global.load_static_weights=false Global.infer_img=doc/imgs_words/en/word_1.jpg
```
Input image:
![](../imgs_words/en/word_1.png)
......
doc/doc_en/training_en.md 0 → 100644
浏览文件 @ f57a6d1b
# MODEL TRAINING
- [1. Basic concepts](#1-basic-concepts)
* [1.1 Learning rate](#11-learning-rate)
* [1.2 Regularization](#12-regularization)
* [1.3 Evaluation indicators](#13-evaluation-indicators-)
- [2. Data and vertical scenes](#2-data-and-vertical-scenes)
* [2.1 Training data](#21-training-data)
* [2.2 Vertical scene](#22-vertical-scene)
* [2.3 Build your own data set](#23-build-your-own-data-set)
* [3. FAQ](#3-faq)
This article will introduce the basic concepts that need to be mastered during model training and the tuning methods during training.
At the same time, it will briefly introduce the components of the PaddleOCR model training data and how to prepare the data finetune model in the vertical scene.
<a name="1-basic-concepts"></a>
# 1. Basic concepts
OCR (Optical Character Recognition) refers to the process of analyzing and recognizing images to obtain text and layout information. It is a typical computer vision task.
It usually consists of two subtasks: text detection and text recognition.
The following parameters need to be paid attention to when tuning the model:
<a name="11-learning-rate"></a>
## 1.1 Learning rate
The learning rate is one of the important hyperparameters for training neural networks. It represents the step length of the gradient moving to the optimal solution of the loss function in each iteration.
A variety of learning rate update strategies are provided in PaddleOCR, which can be modified through configuration files, for example:
```
Optimizer:
...
lr:
name: Piecewise
decay_epochs : [700, 800]
values : [0.001, 0.0001]
warmup_epoch: 5
```
Piecewise stands for piecewise constant attenuation. Different learning rates are specified in different learning stages,
and the learning rate is the same in each stage.
warmup_epoch means that in the first 5 epochs, the learning rate will gradually increase from 0 to base_lr. For all strategies, please refer to the code [learning_rate.py](../../ppocr/optimizer/learning_rate.py).
<a name="12-regularization"></a>
## 1.2 Regularization
Regularization can effectively avoid algorithm overfitting. PaddleOCR provides L1 and L2 regularization methods.
L1 and L2 regularization are the most commonly used regularization methods.
L1 regularization adds a regularization term to the objective function to reduce the sum of absolute values of the parameters;
while in L2 regularization, the purpose of adding a regularization term is to reduce the sum of squared parameters.
The configuration method is as follows:
```
Optimizer:
...
regularizer:
name: L2
factor: 2.0e-05
```
<a name="13-evaluation-indicators-"></a>
## 1.3 Evaluation indicators
(1) Detection stage: First, evaluate according to the IOU of the detection frame and the labeled frame. If the IOU is greater than a certain threshold, it is judged that the detection is accurate. Here, the detection frame and the label frame are different from the general general target detection frame, and they are represented by polygons. Detection accuracy: the percentage of the correct detection frame number in all detection frames is mainly used to judge the detection index. Detection recall rate: the percentage of correct detection frames in all marked frames, which is mainly an indicator of missed detection.
(2) Recognition stage: Character recognition accuracy, that is, the ratio of correctly recognized text lines to the number of marked text lines. Only the entire line of text recognition pairs can be regarded as correct recognition.
(3) End-to-end statistics: End-to-end recall rate: accurately detect and correctly identify the proportion of text lines in all labeled text lines; End-to-end accuracy rate: accurately detect and correctly identify the number of text lines in the detected text lines The standard for accurate detection is that the IOU of the detection box and the labeled box is greater than a certain threshold, and the text in the correctly identified detection box is the same as the labeled text.
<a name="2-data-and-vertical-scenes"></a>
# 2. Data and vertical scenes
<a name="21-training-data"></a>
## 2.1 Training data
The current open source models, data sets and magnitudes are as follows:
- Detection:
- English data set, ICDAR2015
- Chinese data set, LSVT street view data set training data 3w pictures
- Identification:
- English data set, MJSynth and SynthText synthetic data, the data volume is tens of millions.
- Chinese data set, LSVT street view data set crops the image according to the truth value, and performs position calibration, a total of 30w images. In addition, based on the LSVT corpus, 500w of synthesized data.
- Small language data set, using different corpora and fonts, respectively generated 100w synthetic data set, and using ICDAR-MLT as the verification set.
Among them, the public data sets are all open source, users can search and download by themselves, or refer to [Chinese data set](./datasets.md), synthetic data is not open source, users can use open source synthesis tools to synthesize by themselves. Synthesis tools include [text_renderer](https://github.com/Sanster/text_renderer), [SynthText](https://github.com/ankush-me/SynthText), [TextRecognitionDataGenerator](https://github.com/Belval/TextRecognitionDataGenerator) etc.
<a name="22-vertical-scene"></a>
## 2.2 Vertical scene
PaddleOCR mainly focuses on general OCR. If you have vertical requirements, you can use PaddleOCR + vertical data to train yourself;
If there is a lack of labeled data, or if you do not want to invest in research and development costs, it is recommended to directly call the open API, which covers some of the more common vertical categories.
<a name="23-build-your-own-data-set"></a>
## 2.3 Build your own data set
There are several experiences for reference when constructing the data set:
(1) The amount of data in the training set:
a. The data required for detection is relatively small. For Fine-tune based on the PaddleOCR model, 500 sheets are generally required to achieve good results.
b. Recognition is divided into English and Chinese. Generally, English scenarios require hundreds of thousands of data to achieve good results, while Chinese requires several million or more.
(2) When the amount of training data is small, you can try the following three ways to get more data:
a. Manually collect more training data, the most direct and effective way.
b. Basic image processing or transformation based on PIL and opencv. For example, the three modules of ImageFont, Image, ImageDraw in PIL write text into the background, opencv's rotating affine transformation, Gaussian filtering and so on.
c. Use data generation algorithms to synthesize data, such as algorithms such as pix2pix.
<a name="3-faq"></a>
# 3. FAQ
**Q**: How to choose a suitable network input shape when training CRNN recognition?
A: The general height is 32, the longest width is selected, there are two methods:
(1) Calculate the aspect ratio distribution of training sample images. The selection of the maximum aspect ratio considers 80% of the training samples.
(2) Count the number of texts in training samples. The selection of the longest number of characters considers the training sample that satisfies 80%. Then the aspect ratio of Chinese characters is approximately considered to be 1, and that of English is 3:1, and the longest width is estimated.
**Q**: During the recognition training, the accuracy of the training set has reached 90, but the accuracy of the verification set has been kept at 70, what should I do?
A: If the accuracy of the training set is 90 and the test set is more than 70, it should be over-fitting. There are two methods to try:
(1) Add more augmentation methods or increase the [probability] of augmented prob (https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/ppocr/data/imaug/rec_img_aug.py#L341), The default is 0.4.
(2) Increase the [l2 dcay value] of the system (https://github.com/PaddlePaddle/PaddleOCR/blob/a501603d54ff5513fc4fc760319472e59da25424/configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml#L47)
**Q**: When the recognition model is trained, loss can drop normally, but acc is always 0
A: It is normal for the acc to be 0 at the beginning of the recognition model training, and the indicator will come up after a longer training period.
doc/doc_en/update_en.md
浏览文件 @ f57a6d1b
# RECENT UPDATES
- 2021.9.7 release PaddleOCR v2.3, [PP-OCRv2](#PP-OCRv2) is proposed. The CPU inference speed of PP-OCRv2 is 220% higher than that of PP-OCR server. The F-score of PP-OCRv2 is 7% higher than that of PP-OCR mobile.
- 2021.8.3 released PaddleOCR v2.2, add a new structured documents analysis toolkit, i.e., [PP-Structure](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/ppstructure/README.md), support layout analysis and table recognition (One-key to export chart images to Excel files).
- 2021.4.8 release end-to-end text recognition algorithm [PGNet](https://www.aaai.org/AAAI21Papers/AAAI-2885.WangP.pdf) which is published in AAAI 2021. Find tutorial [here](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/pgnet_en.md)；release multi language recognition [models](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md), support more than 80 languages recognition; especically, the performance of [English recognition model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/models_list_en.md#English) is Optimized.
- 2021.1.21 update more than 25+ multilingual recognition models [models list](./doc/doc_en/models_list_en.md), including：English, Chinese, German, French, Japanese，Spanish，Portuguese Russia Arabic and so on. Models for more languages will continue to be updated [Develop Plan](https://github.com/PaddlePaddle/PaddleOCR/issues/1048).
- 2020.12.15 update Data synthesis tool, i.e., [Style-Text](../../StyleText/README.md)，easy to synthesize a large number of images which are similar to the target scene image.
- 2020.11.25 Update a new data annotation tool, i.e., [PPOCRLabel](../../PPOCRLabel/README.md), which is helpful to improve the labeling efficiency. Moreover, the labeling results can be used in training of the PP-OCR system directly.
- 2020.9.22 Update the PP-OCR technical article, https://arxiv.org/abs/2009.09941
......
doc/doc_en/visualization_en.md
浏览文件 @ f57a6d1b
# Visualization
<a name="PP-OCRv2"></a>
## PP-OCRv2
<img src="../imgs_results/PP-OCRv2/PP-OCRv2-pic001.jpg" width="800">
<img src="../imgs_results/PP-OCRv2/PP-OCRv2-pic002.jpg" width="800">
<img src="../imgs_results/PP-OCRv2/PP-OCRv2-pic003.jpg" width="800">
<a name="ppocr_server_2.0"></a>
## ch_ppocr_server_2.0
......
doc/doc_en/whl_en.md
浏览文件 @ f57a6d1b
# paddleocr package
# Paddleocr Package
## 1 Get started quickly
### 1.1 install package
......
doc/joinus.PNG
查看替换文件 @ df4a2f6a 浏览文件 @ f57a6d1b
doc/joinus.PNG

184.8 KB | W: | H:

doc/joinus.PNG

188.4 KB | W: | H:

doc/joinus.PNG
doc/joinus.PNG
doc/joinus.PNG
doc/joinus.PNG
  • 2-up
  • Swipe
  • Onion skin
paddleocr.py
浏览文件 @ f57a6d1b
......@@ -33,104 +33,141 @@ from tools.infer.utility import draw_ocr, str2bool
from ppstructure.utility import init_args, draw_structure_result
from ppstructure.predict_system import OCRSystem, save_structure_res
__all__ = ['PaddleOCR', 'PPStructure', 'draw_ocr', 'draw_structure_result', 'save_structure_res','download_with_progressbar']
model_urls = {
'det': {
'ch':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar',
'en':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_ppocr_mobile_v2.0_det_infer.tar',
'structure': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar'
__all__ = [
'PaddleOCR', 'PPStructure', 'draw_ocr', 'draw_structure_result',
'save_structure_res', 'download_with_progressbar'
]
SUPPORT_DET_MODEL = ['DB']
VERSION = '2.2.1'
SUPPORT_REC_MODEL = ['CRNN']
BASE_DIR = os.path.expanduser("~/.paddleocr/")
DEFAULT_MODEL_VERSION = '2.0'
MODEL_URLS = {
'2.1': {
'det': {
'ch': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_det_infer.tar',
},
},
'rec': {
'ch': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.1/chinese/ch_ppocr_mobile_v2.1_rec_infer.tar',
'dict_path': './ppocr/utils/ppocr_keys_v1.txt'
}
}
},
'rec': {
'ch': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/ppocr_keys_v1.txt'
'2.0': {
'det': {
'ch': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar',
},
'en': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_ppocr_mobile_v2.0_det_infer.tar',
},
'structure': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar'
}
},
'en': {
'url':
'rec': {
'ch': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/ppocr_keys_v1.txt'
},
'en': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_number_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/en_dict.txt'
},
'french': {
'url':
'dict_path': './ppocr/utils/en_dict.txt'
},
'french': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/french_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/french_dict.txt'
},
'german': {
'url':
'dict_path': './ppocr/utils/dict/french_dict.txt'
},
'german': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/german_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/german_dict.txt'
},
'korean': {
'url':
'dict_path': './ppocr/utils/dict/german_dict.txt'
},
'korean': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/korean_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/korean_dict.txt'
},
'japan': {
'url':
'dict_path': './ppocr/utils/dict/korean_dict.txt'
},
'japan': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/japan_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/japan_dict.txt'
},
'chinese_cht': {
'url':
'dict_path': './ppocr/utils/dict/japan_dict.txt'
},
'chinese_cht': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/chinese_cht_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/chinese_cht_dict.txt'
},
'ta': {
'url':
'dict_path': './ppocr/utils/dict/chinese_cht_dict.txt'
},
'ta': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/ta_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/ta_dict.txt'
},
'te': {
'url':
'dict_path': './ppocr/utils/dict/ta_dict.txt'
},
'te': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/te_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/te_dict.txt'
},
'ka': {
'url':
'dict_path': './ppocr/utils/dict/te_dict.txt'
},
'ka': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/ka_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/ka_dict.txt'
},
'latin': {
'url':
'dict_path': './ppocr/utils/dict/ka_dict.txt'
},
'latin': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/latin_ppocr_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/latin_dict.txt'
},
'arabic': {
'url':
'dict_path': './ppocr/utils/dict/latin_dict.txt'
},
'arabic': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/arabic_ppocr_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/arabic_dict.txt'
},
'cyrillic': {
'url':
'dict_path': './ppocr/utils/dict/arabic_dict.txt'
},
'cyrillic': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/cyrillic_ppocr_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/cyrillic_dict.txt'
},
'devanagari': {
'url':
'dict_path': './ppocr/utils/dict/cyrillic_dict.txt'
},
'devanagari': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/devanagari_ppocr_mobile_v2.0_rec_infer.tar',
'dict_path': './ppocr/utils/dict/devanagari_dict.txt'
'dict_path': './ppocr/utils/dict/devanagari_dict.txt'
},
'structure': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar',
'dict_path': 'ppocr/utils/dict/table_dict.txt'
}
},
'cls': {
'ch': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar',
}
},
'structure': {
'url': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar',
'dict_path': 'ppocr/utils/dict/table_dict.txt'
'table': {
'en': {
'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar',
'dict_path': 'ppocr/utils/dict/table_structure_dict.txt'
}
}
},
'cls': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar',
'table': {
'url': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar',
'dict_path': 'ppocr/utils/dict/table_structure_dict.txt'
}
}
SUPPORT_DET_MODEL = ['DB']
VERSION = '2.2.0.1'
SUPPORT_REC_MODEL = ['CRNN']
BASE_DIR = os.path.expanduser("~/.paddleocr/")
def parse_args(mMain=True):
import argparse
......@@ -140,6 +177,7 @@ def parse_args(mMain=True):
parser.add_argument("--det", type=str2bool, default=True)
parser.add_argument("--rec", type=str2bool, default=True)
parser.add_argument("--type", type=str, default='ocr')
parser.add_argument("--version", type=str, default='2.1')
for action in parser._actions:
if action.dest in ['rec_char_dict_path', 'table_char_dict_path']:
......@@ -155,19 +193,19 @@ def parse_args(mMain=True):
def parse_lang(lang):
latin_lang = [
'af', 'az', 'bs', 'cs', 'cy', 'da', 'de', 'es', 'et', 'fr', 'ga',
'hr', 'hu', 'id', 'is', 'it', 'ku', 'la', 'lt', 'lv', 'mi', 'ms',
'mt', 'nl', 'no', 'oc', 'pi', 'pl', 'pt', 'ro', 'rs_latin', 'sk',
'sl', 'sq', 'sv', 'sw', 'tl', 'tr', 'uz', 'vi'
'af', 'az', 'bs', 'cs', 'cy', 'da', 'de', 'es', 'et', 'fr', 'ga', 'hr',
'hu', 'id', 'is', 'it', 'ku', 'la', 'lt', 'lv', 'mi', 'ms', 'mt', 'nl',
'no', 'oc', 'pi', 'pl', 'pt', 'ro', 'rs_latin', 'sk', 'sl', 'sq', 'sv',
'sw', 'tl', 'tr', 'uz', 'vi'
]
arabic_lang = ['ar', 'fa', 'ug', 'ur']
cyrillic_lang = [
'ru', 'rs_cyrillic', 'be', 'bg', 'uk', 'mn', 'abq', 'ady', 'kbd',
'ava', 'dar', 'inh', 'che', 'lbe', 'lez', 'tab'
'ru', 'rs_cyrillic', 'be', 'bg', 'uk', 'mn', 'abq', 'ady', 'kbd', 'ava',
'dar', 'inh', 'che', 'lbe', 'lez', 'tab'
]
devanagari_lang = [
'hi', 'mr', 'ne', 'bh', 'mai', 'ang', 'bho', 'mah', 'sck', 'new',
'gom', 'sa', 'bgc'
'hi', 'mr', 'ne', 'bh', 'mai', 'ang', 'bho', 'mah', 'sck', 'new', 'gom',
'sa', 'bgc'
]
if lang in latin_lang:
lang = "latin"
......@@ -177,9 +215,9 @@ def parse_lang(lang):
lang = "cyrillic"
elif lang in devanagari_lang:
lang = "devanagari"
assert lang in model_urls[
assert lang in MODEL_URLS[DEFAULT_MODEL_VERSION][
'rec'], 'param lang must in {}, but got {}'.format(
model_urls['rec'].keys(), lang)
MODEL_URLS[DEFAULT_MODEL_VERSION]['rec'].keys(), lang)
if lang == "ch":
det_lang = "ch"
elif lang == 'structure':
......@@ -189,6 +227,35 @@ def parse_lang(lang):
return lang, det_lang
def get_model_config(version, model_type, lang):
if version not in MODEL_URLS:
logger.warning('version {} not in {}, use version {} instead'.format(
version, MODEL_URLS.keys(), DEFAULT_MODEL_VERSION))
version = DEFAULT_MODEL_VERSION
if model_type not in MODEL_URLS[version]:
if model_type in MODEL_URLS[DEFAULT_MODEL_VERSION]:
logger.warning(
'version {} not support {} models, use version {} instead'.
format(version, model_type, DEFAULT_MODEL_VERSION))
version = DEFAULT_MODEL_VERSION
else:
logger.error('{} models is not support, we only support {}'.format(
model_type, MODEL_URLS[DEFAULT_MODEL_VERSION].keys()))
sys.exit(-1)
if lang not in MODEL_URLS[version][model_type]:
if lang in MODEL_URLS[DEFAULT_MODEL_VERSION][model_type]:
logger.warning('lang {} is not support in {}, use {} instead'.
format(lang, version, DEFAULT_MODEL_VERSION))
version = DEFAULT_MODEL_VERSION
else:
logger.error(
'lang {} is not support, we only support {} for {} models'.
format(lang, MODEL_URLS[DEFAULT_MODEL_VERSION][model_type].keys(
), model_type))
sys.exit(-1)
return MODEL_URLS[version][model_type][lang]
class PaddleOCR(predict_system.TextSystem):
def __init__(self, **kwargs):
"""
......@@ -204,15 +271,21 @@ class PaddleOCR(predict_system.TextSystem):
lang, det_lang = parse_lang(params.lang)
# init model dir
params.det_model_dir, det_url = confirm_model_dir_url(params.det_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'det', det_lang),
model_urls['det'][det_lang])
params.rec_model_dir, rec_url = confirm_model_dir_url(params.rec_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'rec', lang),
model_urls['rec'][lang]['url'])
params.cls_model_dir, cls_url = confirm_model_dir_url(params.cls_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'cls'),
model_urls['cls'])
det_model_config = get_model_config(params.version, 'det', det_lang)
params.det_model_dir, det_url = confirm_model_dir_url(
params.det_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'det', det_lang),
det_model_config['url'])
rec_model_config = get_model_config(params.version, 'rec', lang)
params.rec_model_dir, rec_url = confirm_model_dir_url(
params.rec_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'rec', lang),
rec_model_config['url'])
cls_model_config = get_model_config(params.version, 'cls', 'ch')
params.cls_model_dir, cls_url = confirm_model_dir_url(
params.cls_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'cls'),
cls_model_config['url'])
# download model
maybe_download(params.det_model_dir, det_url)
maybe_download(params.rec_model_dir, rec_url)
......@@ -226,7 +299,8 @@ class PaddleOCR(predict_system.TextSystem):
sys.exit(0)
if params.rec_char_dict_path is None:
params.rec_char_dict_path = str(Path(__file__).parent / model_urls['rec'][lang]['dict_path'])
params.rec_char_dict_path = str(
Path(__file__).parent / rec_model_config['dict_path'])
print(params)
# init det_model and rec_model
......@@ -293,24 +367,32 @@ class PPStructure(OCRSystem):
lang, det_lang = parse_lang(params.lang)
# init model dir
params.det_model_dir, det_url = confirm_model_dir_url(params.det_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'det', det_lang),
model_urls['det'][det_lang])
params.rec_model_dir, rec_url = confirm_model_dir_url(params.rec_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'rec', lang),
model_urls['rec'][lang]['url'])
params.table_model_dir, table_url = confirm_model_dir_url(params.table_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'table'),
model_urls['table']['url'])
det_model_config = get_model_config(params.version, 'det', det_lang)
params.det_model_dir, det_url = confirm_model_dir_url(
params.det_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'det', det_lang),
det_model_config['url'])
rec_model_config = get_model_config(params.version, 'rec', lang)
params.rec_model_dir, rec_url = confirm_model_dir_url(
params.rec_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'rec', lang),
rec_model_config['url'])
table_model_config = get_model_config(params.version, 'table', 'en')
params.table_model_dir, table_url = confirm_model_dir_url(
params.table_model_dir,
os.path.join(BASE_DIR, VERSION, 'ocr', 'table'),
table_model_config['url'])
# download model
maybe_download(params.det_model_dir, det_url)
maybe_download(params.rec_model_dir, rec_url)
maybe_download(params.table_model_dir, table_url)
if params.rec_char_dict_path is None:
params.rec_char_dict_path = str(Path(__file__).parent / model_urls['rec'][lang]['dict_path'])
params.rec_char_dict_path = str(
Path(__file__).parent / rec_model_config['dict_path'])
if params.table_char_dict_path is None:
params.table_char_dict_path = str(Path(__file__).parent / model_urls['table']['dict_path'])
params.table_char_dict_path = str(
Path(__file__).parent / table_model_config['dict_path'])
print(params)
super().__init__(params)
......@@ -374,4 +456,3 @@ def main():
for item in result:
item.pop('img')
logger.info(item)
ppstructure/README.md
浏览文件 @ f57a6d1b
......@@ -124,8 +124,6 @@ Most of the parameters are consistent with the paddleocr whl package, see [doc o
After running, each image will have a directory with the same name under the directory specified in the output field. Each table in the picture will be stored as an excel and figure area will be cropped and saved, the excel and image file name will be the coordinates of the table in the image.
## 4. PP-Structure Pipeline
the process is as follows
![pipeline](../doc/table/pipeline_en.jpg)
In PP-Structure, the image will be analyzed by layoutparser first. In the layout analysis, the area in the image will be classified, including **text, title, image, list and table** 5 categories. For the first 4 types of areas, directly use the PP-OCR to complete the text detection and recognition. The table area will be converted to an excel file of the same table style via Table OCR.
......
tests/compare_results.py 0 → 100644
浏览文件 @ f57a6d1b
import numpy as np
import os
import subprocess
import json
import argparse
import glob
def init_args():
parser = argparse.ArgumentParser()
# params for testing assert allclose
parser.add_argument("--atol", type=float, default=1e-3)
parser.add_argument("--rtol", type=float, default=1e-3)
parser.add_argument("--gt_file", type=str, default="")
parser.add_argument("--log_file", type=str, default="")
parser.add_argument("--precision", type=str, default="fp32")
return parser
def parse_args():
parser = init_args()
return parser.parse_args()
def run_shell_command(cmd):
p = subprocess.Popen(
cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True)
out, err = p.communicate()
if p.returncode == 0:
return out.decode('utf-8')
else:
return None
def parser_results_from_log_by_name(log_path, names_list):
if not os.path.exists(log_path):
raise ValueError("The log file {} does not exists!".format(log_path))
if names_list is None or len(names_list) < 1:
return []
parser_results = {}
for name in names_list:
cmd = "grep {} {}".format(name, log_path)
outs = run_shell_command(cmd)
outs = outs.split("\n")[0]
result = outs.split("{}".format(name))[-1]
result = json.loads(result)
parser_results[name] = result
return parser_results
def load_gt_from_file(gt_file):
if not os.path.exists(gt_file):
raise ValueError("The log file {} does not exists!".format(gt_file))
with open(gt_file, 'r') as f:
data = f.readlines()
f.close()
parser_gt = {}
for line in data:
image_name, result = line.strip("\n").split("\t")
result = json.loads(result)
parser_gt[image_name] = result
return parser_gt
def load_gt_from_txts(gt_file):
gt_list = glob.glob(gt_file)
gt_collection = {}
for gt_f in gt_list:
gt_dict = load_gt_from_file(gt_f)
basename = os.path.basename(gt_f)
if "fp32" in basename:
gt_collection["fp32"] = [gt_dict, gt_f]
elif "fp16" in basename:
gt_collection["fp16"] = [gt_dict, gt_f]
elif "int8" in basename:
gt_collection["int8"] = [gt_dict, gt_f]
else:
continue
return gt_collection
def collect_predict_from_logs(log_path, key_list):
log_list = glob.glob(log_path)
pred_collection = {}
for log_f in log_list:
pred_dict = parser_results_from_log_by_name(log_f, key_list)
key = os.path.basename(log_f)
pred_collection[key] = pred_dict
return pred_collection
def testing_assert_allclose(dict_x, dict_y, atol=1e-7, rtol=1e-7):
for k in dict_x:
np.testing.assert_allclose(
np.array(dict_x[k]), np.array(dict_y[k]), atol=atol, rtol=rtol)
if __name__ == "__main__":
# Usage:
# python3.7 tests/compare_results.py --gt_file=./tests/results/*.txt --log_file=./tests/output/infer_*.log
args = parse_args()
gt_collection = load_gt_from_txts(args.gt_file)
key_list = gt_collection["fp32"][0].keys()
pred_collection = collect_predict_from_logs(args.log_file, key_list)
for filename in pred_collection.keys():
if "fp32" in filename:
gt_dict, gt_filename = gt_collection["fp32"]
elif "fp16" in filename:
gt_dict, gt_filename = gt_collection["fp16"]
elif "int8" in filename:
gt_dict, gt_filename = gt_collection["int8"]
else:
continue
pred_dict = pred_collection[filename]
try:
testing_assert_allclose(
gt_dict, pred_dict, atol=args.atol, rtol=args.rtol)
print(
"Assert allclose passed! The results of {} and {} are consistent!".
format(filename, gt_filename))
except Exception as E:
print(E)
raise ValueError(
"The results of {} and the results of {} are inconsistent!".
format(filename, gt_filename))
tests/results/det_results_gpu_fp32.txt 0 → 100644
浏览文件 @ f57a6d1b
00008790.jpg [[[209, 406], [280, 406], [280, 419], [209, 419]], [[60, 398], [105, 398], [105, 411], [60, 411]], [[198, 389], [291, 389], [291, 402], [198, 402]], [[162, 391], [173, 391], [173, 401], [162, 401]], [[35, 380], [133, 380], [133, 393], [35, 393]], [[199, 371], [292, 371], [292, 384], [199, 384]], [[218, 310], [272, 310], [272, 324], [218, 324]], [[162, 305], [172, 305], [172, 314], [162, 314]], [[371, 302], [436, 302], [436, 316], [371, 316]], [[31, 302], [134, 301], [134, 315], [31, 316]], [[223, 292], [269, 292], [269, 306], [223, 306]], [[60, 225], [104, 225], [104, 236], [60, 236]], [[218, 223], [272, 223], [272, 237], [218, 237]], [[162, 219], [173, 219], [173, 227], [162, 227]], [[33, 207], [131, 207], [131, 220], [33, 220]], [[223, 206], [269, 206], [269, 220], [223, 220]], [[74, 146], [383, 146], [383, 159], [74, 159]], [[54, 120], [117, 120], [117, 134], [54, 134]], [[74, 51], [296, 51], [296, 65], [74, 65]], [[56, 18], [116, 18], [116, 32], [56, 32]]]
00018946.jpg [[[441, 328], [474, 328], [474, 339], [441, 339]], [[86, 284], [141, 286], [140, 307], [85, 305]], [[302, 279], [377, 279], [377, 297], [302, 297]], [[197, 265], [281, 274], [279, 293], [195, 284]], [[198, 197], [452, 219], [450, 242], [196, 220]], [[343, 182], [376, 182], [376, 192], [343, 192]], [[199, 164], [340, 171], [339, 192], [198, 185]], [[177, 101], [415, 118], [413, 145], [175, 128]]]
00034387.jpg [[[265, 460], [740, 460], [740, 484], [265, 484]], [[348, 417], [420, 417], [420, 443], [348, 443]], [[545, 418], [568, 418], [568, 442], [545, 442]], [[685, 417], [710, 417], [710, 443], [685, 443]], [[175, 415], [226, 415], [226, 443], [175, 443]], [[874, 414], [908, 414], [908, 446], [874, 446]], [[56, 417], [74, 417], [74, 442], [56, 442]], [[856, 373], [925, 373], [925, 400], [856, 400]], [[348, 372], [418, 372], [418, 397], [348, 397]], [[674, 372], [723, 372], [723, 401], [674, 401]], [[539, 373], [570, 373], [570, 400], [539, 400]], [[151, 365], [228, 369], [226, 402], [149, 398]], [[56, 372], [74, 372], [74, 397], [56, 397]], [[857, 329], [925, 329], [925, 355], [857, 355]], [[351, 330], [419, 330], [419, 356], [351, 356]], [[674, 328], [723, 328], [723, 356], [674, 356]], [[541, 329], [570, 329], [570, 357], [541, 357]], [[171, 327], [227, 324], [229, 355], [173, 358]], [[57, 330], [74, 330], [74, 356], [57, 356]], [[298, 327], [316, 327], [316, 334], [298, 334]], [[855, 286], [925, 286], [925, 312], [855, 312]], [[674, 286], [723, 286], [723, 313], [674, 313]], [[346, 286], [426, 283], [427, 313], [347, 316]], [[540, 285], [569, 285], [569, 312], [540, 312]], [[172, 282], [226, 282], [226, 313], [172, 313]], [[56, 287], [73, 287], [73, 312], [56, 312]], [[857, 242], [925, 242], [925, 268], [857, 268]], [[348, 242], [460, 242], [460, 268], [348, 268]], [[156, 242], [227, 242], [227, 269], [156, 269]], [[674, 241], [724, 241], [724, 269], [674, 269]], [[531, 241], [572, 241], [572, 270], [531, 270]], [[56, 242], [74, 242], [74, 268], [56, 268]], [[855, 197], [925, 200], [924, 226], [854, 224]], [[674, 198], [726, 198], [726, 226], [674, 226]], [[344, 200], [434, 195], [436, 223], [346, 228]], [[176, 197], [227, 197], [227, 225], [176, 225]], [[56, 200], [73, 200], [73, 226], [56, 226]], [[527, 194], [576, 194], [576, 226], [527, 226]], [[349, 155], [419, 155], [419, 181], [349, 181]], [[854, 154], [925, 154], [925, 180], [854, 180]], [[176, 154], [226, 154], [226, 183], [176, 183]], [[670, 153], [723, 153], [723, 181], [670, 181]], [[538, 154], [571, 154], [571, 182], [538, 182]], [[56, 156], [74, 156], [74, 182], [56, 182]], [[349, 111], [419, 111], [419, 137], [349, 137]], [[174, 111], [227, 111], [227, 139], [174, 139]], [[546, 113], [564, 113], [564, 137], [546, 137]], [[52, 112], [75, 112], [75, 139], [52, 139]], [[639, 108], [727, 105], [728, 138], [640, 141]], [[817, 103], [927, 110], [925, 139], [815, 132]], [[814, 68], [951, 68], [951, 92], [814, 92]], [[307, 66], [446, 68], [446, 93], [306, 90]], [[673, 67], [723, 67], [723, 93], [673, 93]], [[175, 65], [228, 68], [226, 95], [174, 92]], [[39, 65], [90, 68], [88, 97], [37, 94]], [[528, 65], [580, 65], [580, 94], [528, 94]], [[334, 20], [670, 20], [670, 43], [334, 43]]]
00037951.jpg [[[434, 976], [551, 978], [550, 993], [434, 991]], [[433, 932], [553, 932], [553, 969], [433, 969]], [[30, 522], [98, 522], [98, 545], [30, 545]], [[31, 443], [145, 443], [145, 464], [31, 464]], [[234, 335], [326, 332], [327, 354], [235, 356]], [[124, 252], [436, 252], [436, 284], [124, 284]], [[182, 206], [378, 206], [378, 227], [182, 227]], [[258, 106], [320, 123], [304, 181], [242, 163]], [[28, 65], [33, 65], [33, 71], [28, 71]], [[37, 58], [147, 58], [147, 80], [37, 80]]]
00044782.jpg [[[104, 218], [115, 218], [115, 227], [104, 227]], [[223, 216], [246, 216], [246, 228], [223, 228]], [[163, 216], [182, 216], [182, 229], [163, 229]], [[124, 191], [164, 191], [164, 202], [124, 202]], [[91, 84], [251, 84], [251, 98], [91, 98]], [[73, 63], [278, 63], [278, 78], [73, 78]], [[104, 15], [243, 15], [243, 44], [104, 44]]]
00067516.jpg [[[141, 808], [594, 809], [594, 822], [141, 821]], [[49, 784], [696, 784], [696, 798], [49, 798]], [[579, 751], [667, 751], [667, 764], [579, 764]], [[355, 750], [395, 750], [395, 767], [355, 767]], [[221, 751], [260, 751], [260, 765], [221, 765]], [[477, 750], [501, 750], [501, 768], [477, 768]], [[69, 748], [133, 751], [132, 765], [68, 761]], [[576, 682], [668, 682], [668, 699], [576, 699]], [[476, 682], [518, 682], [518, 700], [476, 700]], [[354, 682], [395, 682], [395, 700], [354, 700]], [[69, 681], [133, 684], [132, 699], [68, 695]], [[220, 679], [243, 682], [241, 700], [218, 697]], [[577, 615], [667, 615], [667, 632], [577, 632]], [[68, 612], [134, 615], [133, 632], [67, 629]], [[476, 614], [500, 614], [500, 633], [476, 633]], [[354, 613], [378, 613], [378, 634], [354, 634]], [[219, 612], [245, 612], [245, 633], [219, 633]], [[578, 547], [667, 547], [667, 564], [578, 564]], [[476, 546], [518, 546], [518, 565], [476, 565]], [[353, 545], [379, 545], [379, 566], [353, 566]], [[219, 545], [245, 545], [245, 566], [219, 566]], [[68, 542], [133, 546], [132, 563], [67, 560]], [[68, 478], [133, 482], [132, 499], [67, 496]], [[586, 481], [664, 481], [664, 497], [586, 497]], [[476, 480], [518, 480], [518, 498], [476, 498]], [[354, 480], [395, 480], [395, 498], [354, 498]], [[219, 479], [245, 479], [245, 500], [219, 500]], [[580, 425], [665, 429], [664, 449], [580, 446]], [[346, 429], [410, 429], [410, 447], [346, 447]], [[68, 426], [150, 429], [149, 449], [67, 447]], [[474, 427], [515, 427], [515, 449], [474, 449]], [[218, 427], [259, 427], [259, 449], [218, 449]], [[283, 398], [478, 399], [478, 419], [283, 418]], [[86, 318], [664, 318], [664, 332], [86, 332]], [[65, 279], [665, 279], [665, 292], [65, 292]], [[458, 210], [584, 210], [584, 224], [458, 224]], [[313, 209], [372, 209], [372, 226], [313, 226]], [[164, 209], [225, 209], [225, 226], [164, 226]], [[505, 151], [539, 151], [539, 166], [505, 166]], [[266, 48], [483, 48], [483, 68], [266, 68]]]
00088568.jpg [[[341, 446], [371, 446], [371, 453], [341, 453]], [[58, 445], [117, 445], [117, 455], [58, 455]], [[552, 433], [571, 433], [571, 440], [552, 440]], [[583, 431], [740, 431], [740, 442], [583, 442]], [[311, 415], [743, 415], [743, 428], [311, 428]], [[311, 377], [736, 377], [736, 390], [311, 390]], [[425, 340], [551, 340], [551, 350], [425, 350]], [[287, 324], [294, 332], [289, 337], [281, 330]], [[276, 294], [348, 296], [347, 311], [276, 309]], [[54, 288], [210, 288], [210, 301], [54, 301]], [[275, 265], [421, 265], [421, 278], [275, 278]], [[56, 264], [248, 264], [248, 277], [56, 277]], [[671, 248], [695, 248], [695, 261], [671, 261]], [[602, 248], [628, 248], [628, 261], [602, 261]], [[533, 248], [557, 248], [557, 261], [533, 261]], [[463, 248], [487, 248], [487, 261], [463, 261]], [[278, 248], [309, 248], [309, 260], [278, 260]], [[55, 240], [142, 240], [142, 254], [55, 254]], [[277, 231], [398, 231], [398, 244], [277, 244]], [[741, 228], [749, 237], [742, 245], [733, 236]], [[665, 230], [700, 230], [700, 244], [665, 244]], [[598, 230], [631, 230], [631, 244], [598, 244]], [[528, 230], [562, 230], [562, 244], [528, 244]], [[459, 230], [492, 230], [492, 244], [459, 244]], [[54, 215], [211, 217], [211, 231], [54, 229]], [[739, 211], [749, 221], [740, 229], [731, 220]], [[663, 214], [704, 214], [704, 228], [663, 228]], [[595, 215], [637, 215], [637, 226], [595, 226]], [[524, 215], [569, 215], [569, 226], [524, 226]], [[454, 215], [495, 215], [495, 226], [454, 226]], [[279, 215], [351, 215], [351, 226], [279, 226]], [[736, 199], [747, 199], [747, 208], [736, 208]], [[668, 197], [700, 197], [700, 208], [668, 208]], [[599, 196], [633, 196], [633, 210], [599, 210]], [[529, 197], [562, 197], [562, 208], [529, 208]], [[461, 197], [491, 197], [491, 208], [461, 208]], [[277, 195], [417, 196], [417, 211], [277, 209]], [[55, 192], [239, 192], [239, 205], [55, 205]], [[665, 181], [703, 181], [703, 192], [665, 192]], [[279, 180], [351, 181], [350, 192], [279, 191]], [[734, 180], [747, 180], [747, 193], [734, 193]], [[597, 180], [634, 180], [634, 191], [597, 191]], [[525, 179], [566, 179], [566, 193], [525, 193]], [[458, 180], [493, 180], [493, 191], [458, 191]], [[55, 170], [142, 170], [142, 184], [55, 184]], [[735, 165], [747, 165], [747, 175], [735, 175]], [[665, 163], [703, 163], [703, 175], [665, 175]], [[598, 163], [634, 163], [634, 175], [598, 175]], [[527, 163], [565, 163], [565, 175], [527, 175]], [[458, 163], [492, 163], [492, 175], [458, 175]], [[279, 162], [398, 162], [398, 176], [279, 176]], [[54, 146], [148, 146], [148, 159], [54, 159]], [[453, 147], [495, 147], [495, 158], [453, 158]], [[731, 143], [748, 146], [745, 161], [728, 158]], [[663, 145], [704, 145], [704, 159], [663, 159]], [[596, 146], [635, 146], [635, 157], [596, 157]], [[522, 145], [566, 142], [567, 157], [523, 159]], [[277, 144], [310, 144], [310, 158], [277, 158]], [[276, 121], [428, 121], [428, 139], [276, 139]], [[52, 120], [232, 121], [232, 139], [52, 138]], [[404, 91], [701, 91], [701, 106], [404, 106]], [[48, 79], [280, 79], [280, 97], [48, 97]], [[325, 69], [744, 70], [744, 84], [325, 83]], [[668, 48], [743, 48], [743, 63], [668, 63]], [[297, 48], [433, 48], [433, 62], [297, 62]]]
00091741.jpg [[[47, 336], [83, 336], [83, 358], [47, 358]], [[99, 211], [257, 211], [257, 230], [99, 230]], [[103, 190], [257, 191], [257, 205], [103, 204]], [[89, 101], [266, 99], [267, 181], [90, 184]], [[94, 48], [262, 55], [260, 114], [91, 107]], [[91, 12], [257, 14], [257, 37], [90, 35]]]
00105313.jpg [[[291, 262], [406, 262], [406, 275], [291, 275]], [[153, 262], [264, 262], [264, 274], [153, 274]], [[11, 258], [73, 261], [72, 274], [11, 272]], [[33, 231], [132, 231], [132, 244], [33, 244]], [[35, 217], [216, 217], [216, 227], [35, 227]], [[33, 200], [146, 200], [146, 213], [33, 213]], [[32, 183], [215, 184], [215, 197], [32, 196]], [[35, 170], [105, 170], [105, 181], [35, 181]], [[35, 155], [124, 155], [124, 164], [35, 164]], [[34, 137], [142, 138], [142, 149], [34, 148]], [[35, 123], [176, 123], [176, 133], [35, 133]], [[33, 106], [176, 106], [176, 119], [33, 119]], [[34, 92], [102, 92], [102, 102], [34, 102]], [[34, 77], [119, 77], [119, 87], [34, 87]], [[32, 60], [120, 60], [120, 73], [32, 73]], [[35, 46], [119, 46], [119, 55], [35, 55]], [[32, 29], [142, 29], [142, 42], [32, 42]], [[25, 12], [147, 12], [147, 24], [25, 24]]]
00134770.jpg [[[388, 646], [456, 646], [456, 656], [388, 656]], [[407, 620], [484, 619], [485, 633], [408, 634]], [[112, 534], [270, 531], [270, 549], [113, 551]], [[111, 502], [443, 497], [443, 514], [112, 519]], [[111, 471], [443, 467], [443, 484], [112, 488]], [[111, 439], [444, 434], [444, 451], [112, 457]], [[111, 409], [442, 405], [442, 421], [112, 425]], [[153, 376], [441, 373], [441, 390], [153, 394]], [[184, 338], [369, 336], [369, 356], [185, 358]], [[75, 98], [515, 104], [513, 218], [74, 212]]]
00145943.jpg [[[394, 248], [746, 279], [731, 449], [379, 418]], [[90, 92], [300, 92], [300, 119], [90, 119]], [[46, 41], [326, 39], [326, 75], [46, 77]]]
00147605.jpg [[[804, 615], [874, 616], [874, 627], [804, 626]], [[516, 607], [784, 605], [784, 628], [516, 629]], [[118, 522], [224, 522], [224, 560], [118, 560]], [[253, 524], [307, 524], [307, 557], [253, 557]], [[715, 501], [900, 505], [900, 538], [714, 534]], [[255, 502], [295, 502], [295, 517], [255, 517]], [[347, 481], [473, 481], [473, 515], [347, 515]], [[252, 484], [295, 484], [295, 499], [252, 499]], [[350, 456], [447, 456], [447, 470], [350, 470]], [[145, 444], [201, 444], [201, 467], [145, 467]], [[728, 371], [878, 371], [878, 420], [728, 420]], [[528, 369], [681, 369], [681, 418], [528, 418]], [[143, 369], [488, 369], [488, 420], [143, 420]], [[744, 315], [871, 315], [871, 336], [744, 336]], [[799, 157], [886, 154], [887, 188], [800, 191]], [[274, 142], [455, 142], [455, 160], [274, 160]], [[738, 116], [894, 119], [893, 157], [737, 153]], [[108, 112], [204, 112], [204, 130], [108, 130]], [[270, 92], [463, 96], [462, 132], [270, 129]]]
00150341.jpg [[[100, 645], [298, 645], [298, 662], [100, 662]], [[115, 617], [288, 617], [288, 631], [115, 631]], [[84, 593], [319, 592], [319, 609], [84, 610]], [[31, 565], [313, 562], [314, 580], [31, 582]], [[444, 560], [461, 560], [461, 569], [444, 569]], [[390, 557], [446, 557], [446, 572], [390, 572]], [[31, 515], [168, 515], [168, 529], [31, 529]], [[33, 490], [110, 490], [110, 504], [33, 504]], [[358, 459], [464, 463], [463, 485], [357, 481]], [[28, 459], [268, 460], [268, 481], [28, 480]], [[339, 439], [421, 444], [421, 460], [338, 455]], [[65, 439], [143, 439], [143, 453], [65, 453]], [[207, 416], [292, 416], [292, 434], [207, 434]], [[319, 408], [441, 413], [440, 438], [318, 433]], [[44, 405], [175, 409], [174, 434], [43, 430]], [[31, 383], [137, 383], [137, 404], [31, 404]]]
00150669.jpg [[[649, 700], [681, 700], [681, 716], [649, 716]], [[517, 685], [549, 685], [549, 720], [517, 720]], [[651, 688], [678, 688], [678, 701], [651, 701]], [[862, 687], [876, 687], [876, 695], [862, 695]], [[922, 675], [938, 675], [938, 685], [922, 685]], [[785, 671], [807, 671], [807, 687], [785, 687]], [[592, 672], [606, 672], [606, 686], [592, 686]], [[722, 679], [732, 669], [742, 678], [731, 688]], [[651, 680], [667, 665], [681, 679], [666, 695]], [[273, 667], [422, 667], [422, 688], [273, 688]], [[136, 666], [203, 666], [203, 688], [136, 688]], [[46, 666], [109, 666], [109, 687], [46, 687]], [[782, 629], [810, 629], [810, 661], [782, 661]], [[645, 627], [685, 627], [685, 665], [645, 665]], [[516, 628], [548, 628], [548, 664], [516, 664]], [[655, 619], [672, 619], [672, 627], [655, 627]], [[598, 617], [605, 624], [599, 629], [592, 622]], [[523, 619], [540, 619], [540, 627], [523, 627]], [[858, 618], [868, 618], [868, 627], [858, 627]], [[727, 618], [735, 618], [735, 627], [727, 627]], [[919, 620], [932, 611], [942, 624], [929, 633]], [[786, 616], [805, 616], [805, 629], [786, 629]], [[373, 604], [420, 604], [420, 619], [373, 619]], [[85, 603], [215, 605], [214, 621], [84, 619]], [[48, 603], [71, 603], [71, 622], [48, 622]], [[788, 561], [806, 561], [806, 572], [788, 572]], [[923, 560], [935, 560], [935, 574], [923, 574]], [[856, 560], [869, 560], [869, 574], [856, 574]], [[62, 554], [410, 554], [410, 568], [62, 568]], [[63, 532], [116, 535], [115, 545], [62, 543]], [[859, 527], [868, 527], [868, 539], [859, 539]], [[925, 526], [934, 526], [934, 540], [925, 540]], [[794, 520], [807, 529], [798, 542], [785, 533]], [[526, 526], [535, 526], [535, 536], [526, 536]], [[262, 513], [395, 513], [395, 526], [262, 526]], [[122, 514], [245, 514], [245, 524], [122, 524]], [[49, 514], [119, 514], [119, 525], [49, 525]], [[755, 492], [828, 492], [828, 507], [755, 507]], [[638, 492], [710, 492], [710, 507], [638, 507]], [[519, 492], [592, 492], [592, 507], [519, 507]], [[85, 450], [123, 450], [123, 461], [85, 461]], [[220, 450], [236, 447], [238, 459], [223, 462]], [[683, 445], [868, 445], [868, 459], [683, 459]], [[562, 445], [666, 445], [666, 459], [562, 459]], [[491, 446], [544, 446], [544, 458], [491, 458]], [[183, 437], [208, 437], [208, 459], [183, 459]], [[52, 431], [72, 438], [64, 462], [44, 455]], [[224, 432], [276, 432], [276, 443], [224, 443]], [[88, 432], [144, 432], [144, 443], [88, 443]], [[506, 383], [616, 382], [616, 397], [506, 398]], [[702, 381], [758, 381], [758, 399], [702, 399]], [[308, 373], [364, 373], [364, 384], [308, 384]], [[92, 373], [167, 373], [167, 384], [92, 384]], [[688, 335], [820, 335], [820, 350], [688, 350]], [[498, 335], [657, 335], [657, 350], [498, 350]], [[208, 316], [244, 316], [244, 331], [208, 331]], [[499, 289], [641, 289], [641, 302], [499, 302]], [[671, 287], [801, 287], [801, 301], [671, 301]], [[670, 241], [816, 241], [816, 255], [670, 255]], [[497, 241], [643, 241], [643, 255], [497, 255]], [[670, 194], [815, 194], [815, 208], [670, 208]], [[498, 194], [643, 194], [643, 208], [498, 208]], [[670, 145], [815, 145], [815, 160], [670, 160]], [[499, 145], [645, 145], [645, 160], [499, 160]], [[489, 103], [546, 103], [546, 120], [489, 120]], [[56, 89], [95, 89], [95, 97], [56, 97]], [[845, 26], [887, 20], [889, 39], [848, 44]], [[26, 20], [700, 20], [700, 37], [26, 37]], [[898, 11], [996, 16], [995, 45], [896, 40]]]
00152568.jpg [[[3, 252], [284, 254], [284, 280], [3, 278]], [[196, 233], [254, 233], [254, 240], [196, 240]], [[49, 229], [90, 229], [90, 240], [49, 240]], [[200, 159], [281, 165], [276, 229], [195, 222]]]
00155628.jpg [[[149, 901], [503, 903], [503, 922], [149, 920]], [[520, 893], [561, 896], [560, 911], [519, 908]], [[61, 885], [81, 885], [81, 894], [61, 894]], [[150, 878], [503, 882], [503, 900], [149, 896]], [[524, 834], [640, 839], [639, 856], [524, 852]], [[70, 834], [185, 835], [185, 853], [69, 852]], [[246, 555], [466, 555], [466, 569], [246, 569]], [[308, 507], [403, 509], [403, 524], [308, 522]], [[244, 482], [459, 484], [459, 502], [244, 500]], [[252, 422], [459, 424], [458, 452], [251, 450]], [[195, 378], [517, 380], [516, 408], [195, 406]], [[474, 194], [624, 196], [624, 210], [473, 208]], [[73, 129], [641, 131], [641, 160], [73, 158]], [[483, 41], [597, 37], [599, 98], [486, 102]], [[68, 25], [135, 16], [139, 43], [72, 52]]]
00173364.jpg [[[8, 178], [57, 178], [57, 200], [8, 200]], [[137, 120], [194, 120], [194, 133], [137, 133]], [[39, 76], [86, 76], [86, 105], [39, 105]], [[249, 20], [310, 20], [310, 36], [249, 36]], [[21, 16], [104, 16], [104, 39], [21, 39]]]
00175503.jpg [[[43, 260], [500, 255], [501, 358], [44, 363]], [[52, 200], [349, 178], [354, 251], [58, 273]]]
00193218.jpg [[[283, 375], [410, 375], [410, 388], [283, 388]], [[172, 375], [221, 375], [221, 389], [172, 389]], [[110, 375], [161, 375], [161, 389], [110, 389]], [[276, 358], [357, 358], [357, 371], [276, 371]], [[171, 359], [220, 359], [220, 370], [171, 370]], [[409, 357], [492, 357], [492, 370], [409, 370]], [[26, 187], [62, 187], [62, 202], [26, 202]], [[501, 185], [557, 185], [557, 199], [501, 199]], [[381, 187], [420, 185], [421, 199], [382, 201]], [[284, 186], [310, 186], [310, 201], [284, 201]], [[174, 186], [196, 186], [196, 201], [174, 201]], [[499, 165], [540, 165], [540, 176], [499, 176]], [[381, 164], [409, 164], [409, 177], [381, 177]], [[262, 163], [302, 163], [302, 177], [262, 177]], [[176, 163], [230, 163], [230, 177], [176, 177]], [[26, 163], [79, 163], [79, 177], [26, 177]], [[387, 140], [488, 140], [488, 153], [387, 153]], [[28, 139], [131, 139], [131, 152], [28, 152]], [[443, 117], [537, 119], [537, 133], [443, 132]], [[346, 119], [405, 119], [405, 130], [346, 130]], [[261, 119], [302, 119], [302, 130], [261, 130]], [[30, 113], [228, 116], [228, 131], [30, 129]], [[131, 91], [394, 94], [394, 109], [131, 105]], [[562, 82], [583, 82], [583, 107], [562, 107]]]
00195033.jpg [[[488, 263], [533, 265], [532, 280], [487, 278]], [[126, 250], [192, 250], [192, 283], [126, 283]], [[338, 249], [362, 249], [362, 266], [338, 266]], [[319, 222], [380, 225], [380, 238], [319, 236]], [[431, 224], [450, 224], [450, 235], [431, 235]], [[365, 203], [538, 203], [538, 216], [365, 216]], [[89, 200], [146, 203], [146, 217], [89, 214]], [[329, 201], [354, 201], [354, 212], [329, 212]], [[371, 181], [449, 181], [449, 194], [371, 194]], [[329, 181], [352, 181], [352, 192], [329, 192]], [[96, 179], [240, 179], [240, 193], [96, 193]], [[456, 162], [555, 162], [555, 175], [456, 175]], [[129, 150], [287, 151], [287, 165], [129, 164]], [[36, 145], [73, 149], [72, 163], [35, 159]], [[527, 146], [552, 146], [552, 155], [527, 155]], [[102, 145], [120, 145], [120, 153], [102, 153]], [[371, 129], [503, 128], [503, 139], [371, 140]], [[99, 126], [193, 126], [193, 139], [99, 139]], [[322, 127], [337, 127], [337, 135], [322, 135]], [[37, 123], [77, 123], [77, 134], [37, 134]], [[324, 106], [337, 106], [337, 115], [324, 115]], [[309, 107], [315, 107], [315, 112], [309, 112]], [[372, 103], [501, 103], [501, 116], [372, 116]], [[349, 105], [360, 105], [360, 114], [349, 114]], [[38, 103], [80, 103], [80, 113], [38, 113]], [[99, 100], [205, 101], [205, 115], [99, 114]], [[306, 90], [317, 90], [317, 97], [306, 97]], [[347, 88], [362, 88], [362, 96], [347, 96]], [[321, 87], [340, 87], [340, 99], [321, 99]], [[358, 84], [513, 82], [513, 95], [358, 97]], [[41, 83], [89, 83], [89, 93], [41, 93]], [[94, 79], [241, 80], [241, 94], [94, 93]], [[313, 66], [394, 66], [394, 79], [313, 79]], [[242, 66], [288, 66], [288, 77], [242, 77]], [[185, 54], [220, 54], [220, 65], [185, 65]], [[469, 48], [547, 48], [547, 61], [469, 61]], [[423, 36], [436, 36], [436, 54], [423, 54]], [[465, 30], [551, 30], [551, 43], [465, 43]], [[207, 21], [329, 23], [328, 41], [207, 39]]]
00208502.jpg [[[247, 566], [282, 566], [282, 573], [247, 573]], [[558, 534], [629, 539], [627, 570], [556, 565]], [[205, 540], [284, 540], [284, 552], [205, 552]], [[143, 513], [189, 513], [189, 525], [143, 525]], [[249, 512], [307, 512], [307, 524], [249, 524]], [[44, 500], [118, 500], [118, 519], [44, 519]], [[467, 491], [556, 491], [556, 508], [467, 508]], [[667, 490], [678, 494], [675, 503], [664, 499]], [[788, 489], [794, 495], [789, 499], [783, 494]], [[726, 491], [737, 491], [737, 501], [726, 501]], [[42, 452], [117, 450], [117, 469], [42, 470]], [[175, 450], [236, 450], [236, 464], [175, 464]], [[614, 407], [638, 407], [638, 422], [614, 422]], [[95, 405], [119, 405], [119, 422], [95, 422]], [[49, 399], [64, 414], [50, 427], [36, 413]], [[209, 401], [226, 401], [226, 415], [209, 415]], [[40, 357], [58, 357], [58, 374], [40, 374]], [[94, 356], [119, 356], [119, 373], [94, 373]], [[188, 341], [246, 339], [247, 361], [189, 364]], [[459, 321], [549, 319], [549, 337], [460, 339]], [[459, 273], [551, 273], [551, 290], [459, 290]], [[563, 272], [735, 269], [735, 286], [564, 289]], [[517, 225], [547, 225], [547, 245], [517, 245]], [[459, 226], [480, 226], [480, 244], [459, 244]], [[621, 187], [673, 187], [673, 201], [621, 201]], [[457, 132], [548, 130], [548, 147], [458, 149]], [[572, 106], [787, 99], [787, 120], [573, 126]], [[122, 48], [290, 48], [290, 97], [122, 97]], [[539, 39], [708, 39], [708, 89], [539, 89]]]
00224225.jpg [[[134, 429], [153, 426], [157, 445], [138, 448]], [[202, 404], [478, 411], [476, 459], [201, 452]], [[205, 230], [469, 230], [469, 390], [205, 390]], [[131, 265], [172, 265], [172, 279], [131, 279]], [[345, 207], [456, 207], [456, 231], [345, 231]], [[199, 189], [346, 196], [344, 239], [197, 232]], [[10, 44], [157, 41], [158, 112], [11, 115]]]
00227746.jpg [[[190, 232], [258, 232], [258, 238], [190, 238]], [[160, 232], [183, 232], [183, 238], [160, 238]], [[123, 232], [150, 232], [150, 238], [123, 238]], [[290, 209], [345, 207], [346, 221], [291, 223]], [[172, 181], [249, 181], [249, 194], [172, 194]], [[143, 178], [165, 180], [162, 208], [140, 206]], [[142, 164], [157, 162], [160, 177], [145, 180]], [[173, 157], [203, 157], [203, 164], [173, 164]], [[200, 154], [347, 154], [347, 167], [200, 167]], [[144, 111], [277, 114], [277, 134], [144, 131]], [[201, 52], [387, 53], [386, 69], [201, 68]], [[139, 47], [191, 45], [192, 62], [140, 64]], [[40, 26], [61, 26], [61, 42], [40, 42]]]
00229605.jpg [[[743, 529], [881, 529], [881, 544], [743, 544]], [[236, 499], [589, 498], [589, 522], [236, 523]], [[6, 498], [227, 498], [227, 522], [6, 522]], [[736, 496], [883, 499], [883, 520], [735, 517]], [[606, 495], [716, 489], [718, 515], [608, 521]], [[4, 245], [863, 230], [864, 288], [5, 303]], [[478, 28], [883, 28], [883, 76], [478, 76]]]
00233011.jpg [[[63, 227], [291, 227], [291, 242], [63, 242]], [[12, 219], [41, 219], [41, 250], [12, 250]], [[61, 177], [119, 177], [119, 195], [61, 195]], [[11, 173], [40, 169], [44, 200], [14, 203]], [[61, 129], [147, 131], [147, 147], [61, 144]], [[12, 124], [43, 124], [43, 154], [12, 154]], [[125, 89], [238, 89], [238, 103], [125, 103]], [[148, 51], [216, 51], [216, 65], [148, 65]], [[258, 46], [353, 50], [352, 69], [257, 65]], [[9, 49], [52, 49], [52, 68], [9, 68]], [[277, 12], [345, 12], [345, 31], [277, 31]], [[28, 11], [73, 11], [73, 31], [28, 31]]]
00233625.jpg [[[375, 397], [632, 399], [632, 443], [375, 440]], [[71, 214], [932, 207], [933, 321], [71, 328]]]
00233634.jpg [[[215, 639], [261, 639], [261, 703], [215, 703]], [[523, 635], [570, 635], [570, 695], [523, 695]], [[643, 523], [682, 523], [682, 568], [643, 568]], [[97, 516], [152, 516], [152, 589], [97, 589]], [[755, 395], [760, 395], [760, 401], [755, 401]], [[26, 395], [32, 395], [32, 400], [26, 400]], [[678, 364], [728, 362], [731, 430], [681, 432]], [[54, 361], [107, 361], [107, 434], [54, 434]], [[78, 208], [155, 208], [155, 280], [78, 280]], [[643, 205], [693, 205], [693, 272], [643, 272]], [[210, 88], [260, 86], [263, 164], [213, 166]], [[363, 48], [426, 45], [430, 115], [367, 118]]]
00234400.jpg [[[446, 421], [738, 421], [738, 438], [446, 438]], [[157, 421], [454, 421], [454, 438], [157, 438]], [[158, 394], [652, 394], [652, 411], [158, 411]], [[40, 391], [127, 391], [127, 412], [40, 412]], [[158, 342], [304, 345], [304, 363], [158, 360]], [[38, 344], [123, 344], [123, 362], [38, 362]], [[520, 295], [703, 295], [703, 314], [520, 314]], [[394, 292], [483, 290], [484, 314], [394, 317]], [[157, 293], [270, 293], [270, 313], [157, 313]], [[37, 293], [125, 293], [125, 313], [37, 313]], [[156, 243], [358, 243], [358, 267], [156, 267]], [[36, 243], [82, 243], [82, 269], [36, 269]], [[29, 152], [158, 152], [158, 175], [29, 175]], [[282, 98], [507, 98], [507, 111], [282, 111]], [[315, 46], [475, 50], [474, 88], [314, 85]], [[518, 51], [663, 53], [662, 67], [517, 65]], [[487, 19], [706, 17], [706, 43], [487, 45]]]
00234883.jpg [[[66, 125], [316, 120], [317, 190], [67, 195]], [[79, 138], [109, 141], [108, 152], [78, 148]], [[72, 120], [120, 120], [120, 130], [72, 130]], [[383, 63], [504, 62], [504, 74], [383, 75]], [[58, 29], [365, 26], [366, 112], [59, 115]], [[387, 28], [501, 26], [501, 45], [387, 47]]]
test_add_0.jpg [[[311, 521], [391, 521], [391, 534], [311, 534]], [[277, 500], [424, 500], [424, 514], [277, 514]], [[261, 446], [437, 446], [437, 459], [261, 459]], [[212, 428], [485, 428], [485, 441], [212, 441]], [[247, 388], [457, 388], [457, 409], [247, 409]], [[222, 328], [474, 328], [474, 372], [222, 372]], [[208, 207], [492, 211], [490, 277], [207, 272]], [[266, 164], [422, 166], [421, 197], [265, 195]], [[18, 20], [201, 18], [201, 43], [18, 45]]]
test_add_1.png []
test_add_10.png [[[157, 124], [186, 124], [186, 172], [157, 172]], [[65, 117], [95, 117], [95, 168], [65, 168]], [[161, 106], [183, 106], [183, 127], [161, 127]], [[69, 100], [94, 100], [94, 128], [69, 128]], [[117, 46], [154, 45], [157, 174], [121, 175]], [[66, 34], [97, 34], [97, 112], [66, 112]]]
test_add_11.jpg [[[1525, 773], [1564, 756], [1575, 780], [1536, 798]], [[1390, 757], [1483, 757], [1483, 791], [1390, 791]], [[1013, 754], [1207, 754], [1207, 800], [1013, 800]], [[685, 755], [875, 755], [875, 796], [685, 796]], [[356, 753], [566, 747], [567, 793], [358, 798]], [[78, 751], [264, 745], [265, 793], [79, 798]], [[602, 647], [1152, 647], [1152, 703], [602, 703]], [[601, 564], [1148, 555], [1149, 611], [602, 620]], [[598, 480], [1066, 472], [1067, 526], [599, 535]], [[598, 393], [1090, 388], [1090, 439], [599, 444]], [[603, 306], [1057, 306], [1057, 357], [603, 357]], [[357, 184], [1517, 184], [1517, 261], [357, 261]], [[60, 43], [257, 37], [259, 83], [61, 89]], [[1305, 41], [1492, 41], [1492, 87], [1305, 87]], [[973, 40], [1171, 34], [1172, 80], [974, 86]], [[670, 40], [862, 34], [864, 80], [671, 86]], [[363, 34], [558, 34], [558, 85], [363, 85]]]
test_add_12.jpg [[[11, 592], [136, 594], [135, 613], [11, 611]], [[109, 521], [907, 526], [907, 569], [109, 565]], [[635, 451], [902, 448], [903, 478], [635, 481]], [[112, 447], [466, 449], [466, 486], [112, 483]], [[582, 306], [680, 304], [681, 348], [583, 351]], [[369, 261], [565, 266], [563, 357], [367, 353]], [[64, 85], [853, 88], [853, 161], [64, 159]]]
test_add_13.jpg [[[68, 94], [117, 97], [116, 114], [67, 111]]]
test_add_14.jpg [[[30, 97], [235, 95], [236, 127], [31, 129]], [[30, 52], [239, 50], [239, 86], [30, 87]]]
test_add_15.jpg [[[141, 253], [353, 253], [353, 266], [141, 266]], [[205, 214], [406, 219], [406, 232], [204, 227]], [[106, 212], [193, 213], [193, 227], [106, 226]], [[154, 156], [286, 161], [286, 174], [154, 170]], [[148, 136], [305, 142], [305, 156], [147, 150]], [[108, 137], [145, 137], [145, 148], [108, 148]], [[108, 102], [275, 109], [275, 125], [107, 117]], [[107, 72], [245, 79], [245, 96], [106, 88]], [[107, 39], [209, 42], [209, 62], [106, 59]]]
test_add_16.jpg [[[398, 842], [408, 842], [408, 852], [398, 852]], [[382, 742], [746, 742], [746, 776], [382, 776]], [[362, 703], [468, 703], [468, 725], [362, 725]], [[1552, 701], [1576, 701], [1576, 746], [1552, 746]], [[1256, 695], [1442, 695], [1442, 721], [1256, 721]], [[1244, 661], [1448, 661], [1448, 687], [1244, 687]], [[386, 645], [668, 645], [668, 679], [386, 679]], [[1226, 625], [1470, 623], [1470, 651], [1226, 653]], [[360, 604], [580, 604], [580, 629], [360, 629]], [[1202, 592], [1494, 592], [1494, 617], [1202, 617]], [[1166, 556], [1530, 556], [1530, 582], [1166, 582]], [[380, 552], [638, 552], [638, 586], [380, 586]], [[356, 502], [516, 502], [516, 536], [356, 536]], [[774, 260], [1124, 260], [1124, 300], [774, 300]], [[374, 210], [504, 210], [504, 300], [374, 300]], [[776, 212], [1088, 217], [1088, 252], [776, 248]]]
test_add_17.jpg [[[321, 255], [393, 258], [392, 271], [320, 269]], [[307, 222], [411, 228], [411, 241], [306, 236]], [[96, 137], [385, 143], [384, 206], [94, 201]], [[72, 95], [399, 103], [398, 124], [71, 117]], [[68, 76], [224, 79], [223, 93], [67, 90]], [[66, 59], [226, 62], [225, 76], [65, 74]]]
test_add_18.jpg [[[466, 788], [715, 790], [715, 813], [466, 811]], [[553, 752], [665, 757], [663, 791], [552, 786]], [[119, 539], [189, 539], [189, 570], [119, 570]], [[116, 473], [674, 486], [673, 528], [115, 516]], [[121, 429], [669, 441], [668, 470], [121, 457]], [[122, 376], [673, 383], [673, 409], [122, 402]], [[556, 262], [675, 264], [675, 278], [556, 277]], [[165, 259], [335, 259], [335, 273], [165, 273]], [[344, 195], [456, 197], [455, 220], [343, 217]], [[309, 175], [490, 175], [490, 190], [309, 190]], [[255, 128], [537, 131], [537, 169], [254, 165]], [[347, 92], [486, 94], [486, 109], [346, 107]], [[285, 41], [567, 49], [566, 82], [284, 74]], [[236, 32], [266, 32], [266, 60], [236, 60]]]
test_add_19.jpg [[[24, 294], [42, 294], [42, 302], [24, 302]], [[64, 293], [105, 293], [105, 303], [64, 303]], [[145, 287], [163, 287], [163, 304], [145, 304]], [[63, 280], [106, 280], [106, 290], [63, 290]], [[9, 281], [26, 281], [26, 288], [9, 288]], [[220, 279], [245, 279], [245, 291], [220, 291]], [[177, 279], [208, 279], [208, 290], [177, 290]], [[23, 279], [51, 279], [51, 290], [23, 290]], [[145, 278], [162, 278], [162, 292], [145, 292]], [[8, 267], [18, 267], [18, 276], [8, 276]], [[221, 265], [243, 265], [243, 277], [221, 277]], [[24, 265], [47, 265], [47, 277], [24, 277]], [[142, 263], [163, 263], [163, 279], [142, 279]], [[218, 252], [249, 252], [249, 265], [218, 265]], [[65, 253], [131, 253], [131, 263], [65, 263]], [[24, 252], [43, 252], [43, 264], [24, 264]], [[8, 253], [18, 253], [18, 262], [8, 262]], [[8, 240], [17, 240], [17, 249], [8, 249]], [[63, 237], [114, 237], [114, 251], [63, 251]], [[25, 236], [47, 239], [45, 251], [23, 249]], [[144, 234], [166, 237], [163, 253], [142, 249]], [[494, 226], [531, 226], [531, 239], [494, 239]], [[335, 226], [354, 226], [354, 237], [335, 237]], [[288, 226], [314, 226], [314, 237], [288, 237]], [[63, 226], [113, 226], [113, 236], [63, 236]], [[7, 227], [17, 227], [17, 234], [7, 234]], [[221, 225], [248, 225], [248, 235], [221, 235]], [[143, 225], [165, 222], [167, 234], [145, 237]], [[24, 224], [48, 224], [48, 238], [24, 238]], [[495, 213], [524, 213], [524, 224], [495, 224]], [[420, 212], [437, 212], [437, 225], [420, 225]], [[336, 212], [398, 212], [398, 223], [336, 223]], [[292, 212], [320, 212], [320, 223], [292, 223]], [[222, 212], [249, 212], [249, 223], [222, 223]], [[145, 212], [166, 212], [166, 223], [145, 223]], [[61, 211], [113, 209], [114, 222], [62, 224]], [[26, 211], [48, 211], [48, 223], [26, 223]], [[337, 199], [383, 199], [383, 209], [337, 209]], [[65, 200], [87, 200], [87, 207], [65, 207]], [[493, 197], [541, 197], [541, 211], [493, 211]], [[445, 202], [455, 196], [462, 206], [452, 212]], [[178, 198], [205, 198], [205, 208], [178, 208]], [[146, 199], [157, 199], [157, 208], [146, 208]], [[32, 194], [43, 204], [33, 214], [22, 203]], [[422, 193], [440, 201], [432, 215], [415, 207]], [[65, 186], [132, 186], [132, 196], [65, 196]], [[337, 185], [399, 185], [399, 196], [337, 196]], [[445, 190], [456, 182], [465, 191], [454, 200]], [[292, 188], [308, 182], [313, 193], [297, 200]], [[220, 183], [255, 183], [255, 197], [220, 197]], [[142, 184], [158, 184], [158, 197], [142, 197]], [[493, 182], [518, 182], [518, 197], [493, 197]], [[425, 180], [437, 191], [427, 202], [414, 190]], [[32, 179], [42, 189], [32, 199], [22, 189]], [[182, 179], [195, 185], [188, 198], [175, 192]], [[335, 172], [400, 169], [400, 183], [336, 185]], [[492, 170], [519, 170], [519, 185], [492, 185]], [[412, 177], [428, 164], [440, 178], [425, 190]], [[293, 171], [315, 171], [315, 185], [293, 185]], [[220, 170], [251, 170], [251, 184], [220, 184]], [[178, 172], [188, 172], [188, 183], [178, 183]], [[64, 172], [125, 170], [125, 181], [64, 182]], [[454, 168], [464, 176], [454, 185], [445, 176]], [[142, 172], [159, 168], [163, 180], [145, 185]], [[30, 165], [43, 174], [34, 186], [20, 177]], [[493, 160], [523, 160], [523, 170], [493, 170]], [[402, 161], [435, 161], [435, 168], [402, 168]], [[335, 159], [401, 159], [401, 169], [335, 169]], [[296, 159], [325, 159], [325, 170], [296, 170]], [[221, 158], [251, 158], [251, 169], [221, 169]], [[174, 161], [183, 156], [190, 167], [181, 172]], [[145, 158], [162, 158], [162, 170], [145, 170]], [[61, 158], [125, 157], [125, 168], [62, 169]], [[20, 161], [33, 154], [40, 167], [28, 174]], [[492, 143], [542, 143], [542, 157], [492, 157]], [[450, 144], [479, 144], [479, 157], [450, 157]], [[335, 143], [439, 143], [439, 156], [335, 156]], [[294, 143], [327, 143], [327, 157], [294, 157]], [[220, 143], [253, 143], [253, 157], [220, 157]], [[178, 145], [187, 145], [187, 156], [178, 156]], [[63, 144], [104, 144], [104, 155], [63, 155]], [[144, 140], [164, 145], [160, 159], [141, 154]], [[31, 137], [44, 149], [31, 162], [17, 149]], [[286, 135], [291, 135], [291, 140], [286, 140]], [[177, 133], [193, 133], [193, 144], [177, 144]], [[336, 132], [388, 132], [388, 141], [336, 141]], [[492, 131], [525, 131], [525, 141], [492, 141]], [[450, 131], [477, 131], [477, 141], [450, 141]], [[292, 131], [321, 131], [321, 141], [292, 141]], [[218, 132], [255, 130], [256, 141], [219, 144]], [[63, 131], [95, 131], [95, 141], [63, 141]], [[417, 130], [437, 130], [437, 141], [417, 141]], [[145, 130], [159, 130], [159, 143], [145, 143]], [[30, 124], [43, 133], [32, 147], [19, 138]], [[493, 118], [535, 118], [535, 129], [493, 129]], [[336, 118], [388, 118], [388, 129], [336, 129]], [[218, 118], [255, 118], [255, 128], [218, 128]], [[451, 117], [478, 117], [478, 129], [451, 129]], [[418, 117], [438, 117], [438, 130], [418, 130]], [[177, 116], [209, 116], [209, 130], [177, 130]], [[145, 117], [162, 117], [162, 130], [145, 130]], [[62, 116], [88, 116], [88, 131], [62, 131]], [[19, 121], [33, 111], [43, 124], [29, 134]], [[491, 107], [523, 107], [523, 113], [491, 113]], [[449, 107], [477, 107], [477, 113], [449, 113]], [[420, 107], [436, 107], [436, 113], [420, 113]], [[295, 107], [319, 107], [319, 114], [295, 114]], [[220, 107], [242, 107], [242, 113], [220, 113]], [[176, 107], [203, 107], [203, 113], [176, 113]], [[145, 107], [161, 107], [161, 114], [145, 114]], [[334, 105], [372, 105], [372, 114], [334, 114]], [[63, 106], [86, 106], [86, 113], [63, 113]], [[483, 89], [522, 89], [522, 99], [483, 99]], [[331, 88], [380, 88], [380, 99], [331, 99]], [[276, 88], [325, 88], [325, 99], [276, 99]], [[214, 88], [246, 88], [246, 99], [214, 99]], [[411, 86], [474, 86], [474, 100], [411, 100]], [[6, 86], [102, 86], [102, 100], [6, 100]], [[415, 66], [461, 66], [461, 77], [415, 77]], [[288, 66], [333, 66], [333, 77], [288, 77]], [[157, 64], [206, 64], [206, 78], [157, 78]], [[416, 48], [523, 49], [523, 63], [415, 62]], [[288, 49], [375, 49], [375, 63], [288, 63]], [[159, 49], [269, 49], [269, 62], [159, 62]], [[24, 53], [36, 46], [45, 59], [33, 67]], [[416, 36], [481, 36], [481, 46], [416, 46]], [[25, 38], [39, 32], [46, 46], [33, 52]], [[157, 34], [205, 34], [205, 47], [157, 47]], [[412, 4], [527, 4], [527, 17], [412, 17]], [[146, 4], [345, 2], [345, 15], [146, 17]]]
test_add_20.jpg [[[31, 346], [605, 346], [605, 370], [31, 370]], [[217, 294], [510, 294], [510, 322], [217, 322]], [[473, 271], [525, 271], [525, 286], [473, 286]], [[220, 267], [287, 267], [287, 286], [220, 286]], [[219, 239], [484, 239], [484, 263], [219, 263]], [[221, 217], [303, 217], [303, 234], [221, 234]], [[402, 192], [417, 192], [417, 205], [402, 205]], [[222, 187], [341, 187], [341, 207], [222, 207]], [[221, 162], [287, 162], [287, 180], [221, 180]], [[375, 122], [475, 124], [475, 146], [375, 143]], [[222, 124], [356, 122], [356, 143], [222, 146]], [[218, 81], [352, 84], [352, 116], [218, 113]], [[440, 35], [605, 35], [605, 60], [440, 60]], [[72, 16], [398, 16], [398, 44], [72, 44]]]
test_add_3.jpg [[[169, 327], [337, 326], [337, 341], [169, 342]], [[170, 288], [307, 290], [307, 312], [170, 310]], [[171, 221], [323, 221], [323, 234], [171, 234]], [[340, 221], [449, 217], [449, 231], [341, 234]], [[169, 201], [372, 201], [372, 214], [169, 214]], [[170, 183], [418, 183], [418, 196], [170, 196]], [[170, 149], [416, 149], [416, 163], [170, 163]], [[171, 119], [418, 119], [418, 140], [171, 140]], [[326, 64], [478, 64], [478, 91], [326, 91]], [[173, 64], [306, 60], [306, 89], [174, 93]]]
test_add_4.png []
test_add_5.png [[[48, 164], [108, 164], [108, 174], [48, 174]], [[52, 121], [169, 121], [169, 134], [52, 134]], [[50, 102], [165, 102], [165, 118], [50, 118]], [[52, 83], [164, 83], [164, 100], [52, 100]], [[51, 68], [166, 68], [166, 84], [51, 84]], [[51, 50], [145, 47], [145, 64], [52, 67]]]
test_add_6.jpg [[[123, 223], [219, 227], [218, 251], [122, 247]], [[172, 172], [186, 186], [172, 200], [158, 186]]]
test_add_7.jpg [[[48, 938], [174, 936], [174, 962], [48, 964]], [[227, 873], [629, 876], [628, 953], [226, 949]], [[56, 745], [638, 745], [638, 790], [56, 790]], [[150, 674], [545, 678], [544, 721], [150, 718]], [[73, 504], [633, 504], [633, 601], [73, 601]], [[59, 270], [655, 279], [652, 441], [56, 432]], [[513, 193], [553, 193], [553, 223], [513, 223]], [[61, 175], [532, 175], [532, 239], [61, 239]], [[533, 178], [642, 178], [642, 236], [533, 236]]]
test_add_8.jpg [[[251, 586], [454, 580], [454, 606], [252, 613]], [[107, 533], [457, 527], [457, 560], [108, 566]], [[336, 494], [384, 494], [384, 507], [336, 507]], [[27, 307], [355, 297], [356, 320], [28, 330]], [[22, 259], [445, 251], [445, 274], [23, 282]], [[78, 209], [445, 205], [445, 225], [78, 229]], [[160, 23], [319, 30], [317, 79], [158, 72]]]
test_add_9.png [[[266, 687], [486, 687], [486, 696], [266, 696]], [[196, 668], [554, 668], [554, 681], [196, 681]], [[154, 596], [597, 596], [597, 606], [154, 606]], [[215, 578], [541, 578], [541, 588], [215, 588]], [[85, 543], [665, 543], [665, 553], [85, 553]], [[96, 522], [653, 522], [653, 535], [96, 535]], [[362, 449], [389, 449], [389, 460], [362, 460]], [[238, 376], [513, 376], [513, 389], [238, 389]], [[177, 356], [574, 356], [574, 368], [177, 368]], [[344, 281], [408, 283], [407, 297], [343, 294]], [[257, 205], [493, 205], [493, 219], [257, 219]]]
tests/results/det_results_gpu_trt_fp16.txt 0 → 100644
浏览文件 @ f57a6d1b
00008790.jpg [[[209, 406], [280, 406], [280, 419], [209, 419]], [[60, 398], [105, 398], [105, 411], [60, 411]], [[198, 389], [291, 389], [291, 402], [198, 402]], [[162, 391], [173, 391], [173, 401], [162, 401]], [[35, 380], [133, 380], [133, 393], [35, 393]], [[199, 371], [292, 371], [292, 384], [199, 384]], [[218, 310], [272, 310], [272, 324], [218, 324]], [[162, 305], [172, 305], [172, 314], [162, 314]], [[371, 302], [436, 302], [436, 316], [371, 316]], [[31, 302], [134, 301], [134, 315], [31, 316]], [[223, 292], [269, 292], [269, 306], [223, 306]], [[60, 225], [104, 225], [104, 236], [60, 236]], [[218, 223], [272, 223], [272, 237], [218, 237]], [[162, 219], [173, 219], [173, 227], [162, 227]], [[33, 207], [131, 207], [131, 220], [33, 220]], [[223, 206], [269, 206], [269, 220], [223, 220]], [[74, 146], [383, 146], [383, 159], [74, 159]], [[54, 120], [117, 120], [117, 134], [54, 134]], [[74, 51], [296, 51], [296, 65], [74, 65]], [[56, 18], [116, 18], [116, 32], [56, 32]]]
00018946.jpg [[[441, 328], [474, 328], [474, 339], [441, 339]], [[86, 284], [141, 286], [140, 307], [85, 305]], [[302, 279], [377, 279], [377, 297], [302, 297]], [[197, 265], [281, 274], [279, 293], [195, 284]], [[198, 197], [452, 219], [450, 242], [196, 220]], [[343, 182], [376, 182], [376, 192], [343, 192]], [[199, 164], [340, 171], [339, 192], [198, 185]], [[177, 101], [415, 118], [413, 145], [175, 128]]]
00034387.jpg [[[265, 460], [740, 460], [740, 484], [265, 484]], [[348, 417], [420, 417], [420, 443], [348, 443]], [[545, 418], [568, 418], [568, 442], [545, 442]], [[685, 417], [710, 417], [710, 443], [685, 443]], [[175, 415], [226, 415], [226, 443], [175, 443]], [[874, 414], [908, 414], [908, 446], [874, 446]], [[56, 417], [74, 417], [74, 442], [56, 442]], [[856, 373], [925, 373], [925, 400], [856, 400]], [[348, 372], [418, 372], [418, 397], [348, 397]], [[674, 372], [723, 372], [723, 401], [674, 401]], [[539, 373], [570, 373], [570, 400], [539, 400]], [[151, 365], [228, 369], [226, 402], [149, 398]], [[56, 372], [74, 372], [74, 397], [56, 397]], [[857, 329], [925, 329], [925, 355], [857, 355]], [[351, 330], [419, 330], [419, 356], [351, 356]], [[674, 328], [723, 328], [723, 356], [674, 356]], [[541, 329], [570, 329], [570, 357], [541, 357]], [[171, 327], [227, 324], [229, 355], [173, 358]], [[57, 330], [74, 330], [74, 356], [57, 356]], [[298, 327], [316, 327], [316, 334], [298, 334]], [[855, 286], [925, 286], [925, 312], [855, 312]], [[674, 286], [723, 286], [723, 313], [674, 313]], [[346, 286], [426, 283], [427, 313], [347, 316]], [[540, 285], [569, 285], [569, 312], [540, 312]], [[172, 282], [226, 282], [226, 313], [172, 313]], [[56, 287], [73, 287], [73, 312], [56, 312]], [[857, 242], [925, 242], [925, 268], [857, 268]], [[348, 242], [460, 242], [460, 268], [348, 268]], [[156, 242], [227, 242], [227, 269], [156, 269]], [[674, 241], [724, 241], [724, 269], [674, 269]], [[531, 241], [572, 241], [572, 270], [531, 270]], [[56, 242], [74, 242], [74, 268], [56, 268]], [[855, 197], [925, 200], [924, 226], [854, 224]], [[674, 198], [726, 198], [726, 226], [674, 226]], [[344, 200], [430, 195], [432, 224], [346, 230]], [[176, 197], [227, 197], [227, 225], [176, 225]], [[56, 200], [73, 200], [73, 226], [56, 226]], [[527, 194], [576, 194], [576, 226], [527, 226]], [[349, 155], [419, 155], [419, 181], [349, 181]], [[854, 154], [925, 154], [925, 180], [854, 180]], [[176, 154], [226, 154], [226, 183], [176, 183]], [[670, 153], [723, 153], [723, 181], [670, 181]], [[538, 154], [571, 154], [571, 182], [538, 182]], [[56, 156], [74, 156], [74, 182], [56, 182]], [[349, 111], [419, 111], [419, 137], [349, 137]], [[174, 111], [227, 111], [227, 139], [174, 139]], [[546, 113], [564, 113], [564, 137], [546, 137]], [[52, 112], [75, 112], [75, 139], [52, 139]], [[639, 108], [727, 105], [728, 138], [640, 141]], [[817, 103], [927, 110], [925, 139], [815, 132]], [[814, 68], [951, 68], [951, 92], [814, 92]], [[307, 66], [446, 68], [446, 93], [306, 90]], [[673, 67], [723, 67], [723, 93], [673, 93]], [[175, 65], [228, 68], [226, 95], [174, 92]], [[39, 65], [90, 68], [88, 97], [37, 94]], [[528, 65], [580, 65], [580, 94], [528, 94]], [[334, 20], [670, 20], [670, 43], [334, 43]]]
00037951.jpg [[[434, 976], [551, 978], [550, 993], [434, 991]], [[433, 932], [553, 932], [553, 969], [433, 969]], [[30, 522], [98, 522], [98, 545], [30, 545]], [[31, 443], [145, 443], [145, 464], [31, 464]], [[234, 335], [326, 332], [327, 354], [235, 356]], [[124, 252], [436, 252], [436, 284], [124, 284]], [[182, 206], [378, 206], [378, 227], [182, 227]], [[258, 106], [320, 123], [304, 181], [242, 163]], [[28, 65], [33, 65], [33, 71], [28, 71]], [[37, 58], [147, 58], [147, 80], [37, 80]]]
00044782.jpg [[[104, 218], [115, 218], [115, 227], [104, 227]], [[223, 216], [246, 216], [246, 228], [223, 228]], [[163, 216], [182, 216], [182, 229], [163, 229]], [[124, 191], [164, 191], [164, 202], [124, 202]], [[91, 84], [251, 84], [251, 98], [91, 98]], [[73, 63], [278, 63], [278, 78], [73, 78]], [[104, 15], [243, 15], [243, 44], [104, 44]]]
00067516.jpg [[[141, 808], [594, 809], [594, 822], [141, 821]], [[49, 784], [696, 784], [696, 798], [49, 798]], [[579, 751], [667, 751], [667, 764], [579, 764]], [[355, 750], [395, 750], [395, 767], [355, 767]], [[221, 751], [260, 751], [260, 765], [221, 765]], [[477, 750], [501, 750], [501, 768], [477, 768]], [[69, 748], [133, 751], [132, 765], [68, 761]], [[576, 682], [668, 682], [668, 699], [576, 699]], [[476, 682], [518, 682], [518, 700], [476, 700]], [[354, 682], [395, 682], [395, 700], [354, 700]], [[69, 681], [133, 684], [132, 699], [68, 695]], [[220, 679], [243, 682], [241, 700], [218, 697]], [[577, 615], [667, 615], [667, 632], [577, 632]], [[68, 612], [134, 615], [133, 632], [67, 629]], [[476, 614], [500, 614], [500, 633], [476, 633]], [[354, 613], [378, 613], [378, 634], [354, 634]], [[219, 612], [245, 612], [245, 633], [219, 633]], [[578, 547], [667, 547], [667, 564], [578, 564]], [[476, 546], [518, 546], [518, 565], [476, 565]], [[353, 545], [379, 545], [379, 566], [353, 566]], [[219, 545], [245, 545], [245, 566], [219, 566]], [[68, 542], [133, 546], [132, 563], [67, 560]], [[68, 478], [133, 482], [132, 499], [67, 496]], [[586, 481], [664, 481], [664, 497], [586, 497]], [[476, 480], [518, 480], [518, 498], [476, 498]], [[354, 480], [395, 480], [395, 498], [354, 498]], [[219, 479], [245, 479], [245, 500], [219, 500]], [[580, 425], [665, 429], [664, 449], [580, 446]], [[346, 429], [410, 429], [410, 447], [346, 447]], [[68, 426], [150, 429], [149, 449], [67, 447]], [[474, 427], [515, 427], [515, 449], [474, 449]], [[218, 427], [259, 427], [259, 449], [218, 449]], [[283, 398], [478, 399], [478, 419], [283, 418]], [[86, 318], [664, 318], [664, 332], [86, 332]], [[65, 279], [665, 279], [665, 292], [65, 292]], [[458, 210], [584, 210], [584, 224], [458, 224]], [[313, 209], [372, 209], [372, 226], [313, 226]], [[164, 209], [225, 209], [225, 226], [164, 226]], [[505, 151], [539, 151], [539, 166], [505, 166]], [[266, 48], [483, 48], [483, 68], [266, 68]]]
00088568.jpg [[[341, 446], [371, 446], [371, 453], [341, 453]], [[58, 445], [117, 445], [117, 455], [58, 455]], [[552, 433], [571, 433], [571, 440], [552, 440]], [[583, 431], [740, 431], [740, 442], [583, 442]], [[311, 415], [743, 415], [743, 428], [311, 428]], [[311, 377], [736, 377], [736, 390], [311, 390]], [[425, 340], [551, 340], [551, 350], [425, 350]], [[287, 324], [294, 332], [289, 337], [281, 330]], [[276, 294], [348, 296], [347, 311], [276, 309]], [[54, 288], [210, 288], [210, 301], [54, 301]], [[275, 265], [421, 265], [421, 278], [275, 278]], [[56, 264], [248, 264], [248, 277], [56, 277]], [[671, 248], [695, 248], [695, 261], [671, 261]], [[602, 248], [628, 248], [628, 261], [602, 261]], [[533, 248], [557, 248], [557, 261], [533, 261]], [[463, 248], [487, 248], [487, 261], [463, 261]], [[278, 248], [309, 248], [309, 260], [278, 260]], [[55, 240], [142, 240], [142, 254], [55, 254]], [[277, 231], [398, 231], [398, 244], [277, 244]], [[741, 228], [749, 237], [742, 245], [733, 236]], [[665, 230], [700, 230], [700, 244], [665, 244]], [[598, 230], [631, 230], [631, 244], [598, 244]], [[528, 230], [562, 230], [562, 244], [528, 244]], [[459, 230], [492, 230], [492, 244], [459, 244]], [[54, 215], [211, 217], [211, 231], [54, 229]], [[739, 211], [749, 221], [740, 229], [731, 220]], [[663, 214], [704, 214], [704, 228], [663, 228]], [[595, 215], [637, 215], [637, 226], [595, 226]], [[524, 215], [568, 215], [568, 226], [524, 226]], [[454, 215], [495, 215], [495, 226], [454, 226]], [[279, 215], [351, 215], [351, 226], [279, 226]], [[736, 199], [747, 199], [747, 208], [736, 208]], [[668, 197], [700, 197], [700, 208], [668, 208]], [[599, 196], [633, 196], [633, 210], [599, 210]], [[529, 197], [562, 197], [562, 208], [529, 208]], [[461, 197], [491, 197], [491, 208], [461, 208]], [[277, 195], [417, 196], [417, 211], [277, 209]], [[55, 192], [239, 192], [239, 205], [55, 205]], [[665, 181], [703, 181], [703, 192], [665, 192]], [[279, 180], [351, 181], [350, 192], [279, 191]], [[734, 180], [747, 180], [747, 193], [734, 193]], [[597, 180], [634, 180], [634, 191], [597, 191]], [[525, 179], [566, 179], [566, 193], [525, 193]], [[458, 180], [493, 180], [493, 191], [458, 191]], [[55, 170], [142, 170], [142, 184], [55, 184]], [[735, 165], [747, 165], [747, 175], [735, 175]], [[665, 163], [703, 163], [703, 175], [665, 175]], [[598, 163], [634, 163], [634, 175], [598, 175]], [[527, 163], [565, 163], [565, 175], [527, 175]], [[458, 163], [492, 163], [492, 175], [458, 175]], [[279, 162], [398, 162], [398, 176], [279, 176]], [[54, 146], [148, 146], [148, 159], [54, 159]], [[453, 147], [495, 147], [495, 158], [453, 158]], [[731, 143], [748, 146], [745, 161], [728, 158]], [[663, 145], [704, 145], [704, 159], [663, 159]], [[596, 146], [635, 146], [635, 157], [596, 157]], [[522, 145], [566, 142], [567, 157], [523, 159]], [[277, 144], [310, 144], [310, 158], [277, 158]], [[276, 121], [428, 121], [428, 139], [276, 139]], [[52, 120], [232, 121], [232, 139], [52, 138]], [[404, 91], [701, 91], [701, 106], [404, 106]], [[48, 79], [280, 79], [280, 97], [48, 97]], [[325, 69], [744, 70], [744, 84], [325, 83]], [[668, 48], [743, 48], [743, 63], [668, 63]], [[297, 48], [433, 48], [433, 62], [297, 62]]]
00091741.jpg [[[47, 336], [83, 336], [83, 358], [47, 358]], [[98, 211], [257, 209], [257, 229], [98, 231]], [[103, 190], [257, 191], [257, 205], [103, 204]], [[89, 101], [266, 99], [267, 181], [90, 184]], [[94, 48], [262, 55], [260, 114], [91, 107]], [[91, 12], [257, 14], [257, 37], [90, 35]]]
00105313.jpg [[[291, 262], [406, 262], [406, 275], [291, 275]], [[153, 262], [264, 262], [264, 274], [153, 274]], [[11, 258], [73, 261], [72, 274], [11, 272]], [[33, 231], [132, 231], [132, 244], [33, 244]], [[35, 217], [216, 217], [216, 227], [35, 227]], [[33, 200], [146, 200], [146, 213], [33, 213]], [[32, 183], [215, 184], [215, 197], [32, 196]], [[35, 170], [105, 170], [105, 181], [35, 181]], [[35, 155], [124, 155], [124, 164], [35, 164]], [[34, 137], [142, 138], [142, 149], [34, 148]], [[35, 123], [176, 123], [176, 133], [35, 133]], [[33, 106], [176, 106], [176, 119], [33, 119]], [[34, 92], [102, 92], [102, 102], [34, 102]], [[34, 77], [119, 77], [119, 87], [34, 87]], [[32, 60], [120, 60], [120, 73], [32, 73]], [[35, 46], [119, 46], [119, 55], [35, 55]], [[32, 29], [142, 29], [142, 42], [32, 42]], [[25, 12], [147, 12], [147, 24], [25, 24]]]
00134770.jpg [[[388, 646], [456, 646], [456, 656], [388, 656]], [[407, 620], [484, 619], [485, 633], [408, 634]], [[112, 534], [270, 531], [270, 549], [113, 551]], [[111, 502], [443, 497], [443, 514], [112, 519]], [[111, 471], [443, 467], [443, 484], [112, 488]], [[111, 439], [444, 434], [444, 451], [112, 457]], [[111, 409], [442, 405], [442, 421], [112, 425]], [[153, 376], [441, 373], [441, 390], [153, 394]], [[184, 338], [369, 336], [369, 356], [185, 358]], [[75, 98], [515, 104], [513, 218], [74, 212]]]
00145943.jpg [[[394, 248], [746, 279], [731, 449], [379, 418]], [[90, 92], [300, 92], [300, 119], [90, 119]], [[46, 41], [326, 39], [326, 75], [46, 77]]]
00147605.jpg [[[805, 616], [874, 616], [874, 627], [805, 627]], [[516, 607], [784, 605], [784, 628], [516, 629]], [[118, 522], [224, 522], [224, 560], [118, 560]], [[253, 524], [307, 524], [307, 557], [253, 557]], [[715, 501], [900, 505], [900, 538], [714, 534]], [[255, 502], [295, 502], [295, 517], [255, 517]], [[347, 481], [473, 481], [473, 515], [347, 515]], [[252, 484], [295, 484], [295, 499], [252, 499]], [[350, 456], [447, 456], [447, 470], [350, 470]], [[145, 444], [201, 444], [201, 467], [145, 467]], [[728, 371], [878, 371], [878, 420], [728, 420]], [[528, 369], [681, 369], [681, 418], [528, 418]], [[143, 369], [488, 369], [488, 420], [143, 420]], [[744, 315], [871, 315], [871, 336], [744, 336]], [[799, 157], [886, 154], [887, 188], [800, 191]], [[274, 142], [455, 142], [455, 160], [274, 160]], [[738, 116], [894, 119], [893, 157], [737, 153]], [[108, 112], [204, 112], [204, 130], [108, 130]], [[270, 92], [463, 96], [462, 132], [270, 129]]]
00150341.jpg [[[100, 644], [297, 644], [297, 661], [100, 661]], [[115, 617], [288, 617], [288, 631], [115, 631]], [[84, 593], [319, 592], [319, 609], [84, 610]], [[31, 565], [313, 562], [314, 580], [31, 582]], [[444, 560], [461, 560], [461, 569], [444, 569]], [[390, 557], [446, 557], [446, 572], [390, 572]], [[31, 515], [168, 515], [168, 529], [31, 529]], [[33, 490], [110, 490], [110, 504], [33, 504]], [[358, 459], [464, 463], [463, 485], [357, 481]], [[28, 459], [268, 460], [268, 481], [28, 480]], [[339, 439], [421, 444], [421, 460], [338, 455]], [[65, 439], [143, 439], [143, 453], [65, 453]], [[207, 416], [292, 416], [292, 434], [207, 434]], [[319, 408], [441, 413], [440, 438], [318, 433]], [[44, 405], [175, 409], [174, 434], [43, 430]], [[31, 383], [137, 383], [137, 404], [31, 404]]]
00150669.jpg [[[649, 700], [681, 700], [681, 716], [649, 716]], [[517, 685], [549, 685], [549, 720], [517, 720]], [[651, 688], [678, 688], [678, 701], [651, 701]], [[862, 687], [876, 687], [876, 695], [862, 695]], [[922, 675], [938, 675], [938, 685], [922, 685]], [[785, 671], [807, 671], [807, 687], [785, 687]], [[592, 672], [606, 672], [606, 686], [592, 686]], [[722, 679], [732, 669], [742, 678], [731, 688]], [[651, 680], [667, 665], [681, 679], [666, 695]], [[273, 667], [422, 667], [422, 688], [273, 688]], [[47, 668], [108, 668], [108, 686], [47, 686]], [[136, 666], [203, 666], [203, 688], [136, 688]], [[782, 629], [810, 629], [810, 661], [782, 661]], [[645, 627], [685, 627], [685, 665], [645, 665]], [[516, 628], [548, 628], [548, 664], [516, 664]], [[655, 619], [672, 619], [672, 627], [655, 627]], [[598, 617], [605, 624], [599, 629], [592, 622]], [[523, 619], [540, 619], [540, 627], [523, 627]], [[858, 618], [868, 618], [868, 627], [858, 627]], [[727, 618], [735, 618], [735, 627], [727, 627]], [[919, 620], [932, 611], [942, 624], [929, 633]], [[786, 616], [805, 616], [805, 629], [786, 629]], [[373, 604], [420, 604], [420, 619], [373, 619]], [[85, 603], [215, 605], [214, 621], [84, 619]], [[48, 603], [71, 603], [71, 622], [48, 622]], [[788, 561], [806, 561], [806, 572], [788, 572]], [[923, 560], [935, 560], [935, 574], [923, 574]], [[856, 560], [869, 560], [869, 574], [856, 574]], [[62, 554], [410, 554], [410, 568], [62, 568]], [[63, 532], [116, 535], [115, 545], [62, 543]], [[859, 527], [868, 527], [868, 539], [859, 539]], [[925, 526], [934, 526], [934, 540], [925, 540]], [[794, 520], [807, 529], [798, 542], [785, 533]], [[526, 526], [535, 526], [535, 536], [526, 536]], [[262, 513], [395, 513], [395, 526], [262, 526]], [[122, 514], [245, 514], [245, 524], [122, 524]], [[49, 514], [119, 514], [119, 525], [49, 525]], [[755, 492], [828, 492], [828, 507], [755, 507]], [[638, 492], [710, 492], [710, 507], [638, 507]], [[519, 492], [592, 492], [592, 507], [519, 507]], [[85, 450], [123, 450], [123, 461], [85, 461]], [[220, 450], [236, 447], [238, 459], [223, 462]], [[683, 445], [868, 445], [868, 459], [683, 459]], [[562, 445], [666, 445], [666, 459], [562, 459]], [[491, 446], [544, 446], [544, 458], [491, 458]], [[183, 437], [208, 437], [208, 459], [183, 459]], [[52, 431], [72, 438], [64, 462], [44, 455]], [[224, 432], [276, 432], [276, 443], [224, 443]], [[88, 432], [144, 432], [144, 443], [88, 443]], [[506, 383], [616, 382], [616, 397], [506, 398]], [[702, 381], [758, 381], [758, 399], [702, 399]], [[308, 373], [364, 373], [364, 384], [308, 384]], [[92, 373], [167, 373], [167, 384], [92, 384]], [[688, 335], [820, 335], [820, 350], [688, 350]], [[498, 335], [657, 335], [657, 350], [498, 350]], [[208, 316], [244, 316], [244, 331], [208, 331]], [[499, 289], [641, 289], [641, 302], [499, 302]], [[671, 287], [801, 287], [801, 301], [671, 301]], [[670, 241], [816, 241], [816, 255], [670, 255]], [[497, 241], [643, 241], [643, 255], [497, 255]], [[670, 194], [815, 194], [815, 208], [670, 208]], [[498, 194], [643, 194], [643, 208], [498, 208]], [[670, 145], [815, 145], [815, 160], [670, 160]], [[499, 145], [645, 145], [645, 160], [499, 160]], [[489, 103], [546, 103], [546, 120], [489, 120]], [[56, 89], [95, 89], [95, 97], [56, 97]], [[845, 26], [887, 20], [889, 39], [848, 44]], [[26, 20], [700, 20], [700, 37], [26, 37]], [[898, 11], [996, 16], [995, 45], [896, 40]]]
00152568.jpg [[[3, 252], [284, 254], [284, 280], [3, 278]], [[196, 233], [254, 233], [254, 240], [196, 240]], [[49, 229], [90, 229], [90, 240], [49, 240]], [[200, 159], [281, 165], [276, 229], [195, 222]]]
00155628.jpg [[[149, 901], [503, 903], [503, 922], [149, 920]], [[520, 893], [561, 896], [560, 911], [519, 908]], [[61, 885], [81, 885], [81, 894], [61, 894]], [[150, 878], [503, 882], [503, 900], [149, 896]], [[524, 834], [640, 839], [639, 856], [524, 852]], [[70, 834], [185, 835], [185, 853], [69, 852]], [[246, 555], [466, 555], [466, 569], [246, 569]], [[308, 507], [403, 509], [403, 524], [308, 522]], [[244, 482], [459, 484], [459, 502], [244, 500]], [[252, 422], [459, 424], [458, 452], [251, 450]], [[195, 378], [517, 380], [516, 408], [195, 406]], [[474, 194], [624, 196], [624, 210], [473, 208]], [[73, 129], [641, 131], [641, 160], [73, 158]], [[483, 41], [597, 37], [599, 98], [486, 102]], [[68, 25], [135, 16], [139, 43], [72, 52]]]
00173364.jpg [[[8, 178], [56, 178], [56, 200], [8, 200]], [[137, 120], [194, 120], [194, 133], [137, 133]], [[39, 76], [86, 76], [86, 105], [39, 105]], [[249, 20], [310, 20], [310, 36], [249, 36]], [[21, 16], [104, 16], [104, 39], [21, 39]]]
00175503.jpg [[[43, 260], [500, 255], [501, 358], [44, 363]], [[52, 200], [349, 178], [354, 251], [58, 273]]]
00193218.jpg [[[283, 375], [410, 375], [410, 388], [283, 388]], [[172, 375], [221, 375], [221, 389], [172, 389]], [[110, 375], [161, 375], [161, 389], [110, 389]], [[276, 358], [357, 358], [357, 371], [276, 371]], [[171, 359], [220, 359], [220, 370], [171, 370]], [[409, 357], [492, 357], [492, 370], [409, 370]], [[26, 187], [62, 187], [62, 202], [26, 202]], [[501, 185], [557, 185], [557, 199], [501, 199]], [[381, 187], [420, 185], [421, 199], [382, 201]], [[284, 186], [310, 186], [310, 201], [284, 201]], [[174, 186], [196, 186], [196, 201], [174, 201]], [[499, 165], [540, 165], [540, 176], [499, 176]], [[381, 164], [409, 164], [409, 177], [381, 177]], [[262, 163], [302, 163], [302, 177], [262, 177]], [[176, 163], [230, 163], [230, 177], [176, 177]], [[26, 163], [79, 163], [79, 177], [26, 177]], [[387, 140], [488, 140], [488, 153], [387, 153]], [[28, 139], [131, 139], [131, 152], [28, 152]], [[443, 117], [537, 119], [537, 133], [443, 132]], [[346, 119], [405, 119], [405, 130], [346, 130]], [[261, 119], [303, 119], [303, 130], [261, 130]], [[30, 113], [228, 116], [228, 131], [30, 129]], [[131, 91], [394, 94], [394, 109], [131, 105]], [[562, 82], [583, 82], [583, 107], [562, 107]]]
00195033.jpg [[[488, 263], [533, 265], [532, 280], [487, 278]], [[126, 250], [192, 250], [192, 283], [126, 283]], [[338, 249], [362, 249], [362, 266], [338, 266]], [[319, 222], [380, 225], [380, 238], [319, 236]], [[431, 224], [450, 224], [450, 235], [431, 235]], [[365, 203], [538, 203], [538, 216], [365, 216]], [[89, 200], [146, 203], [146, 217], [89, 214]], [[329, 201], [354, 201], [354, 212], [329, 212]], [[371, 181], [449, 181], [449, 194], [371, 194]], [[329, 181], [352, 181], [352, 192], [329, 192]], [[96, 179], [240, 179], [240, 193], [96, 193]], [[456, 162], [555, 162], [555, 175], [456, 175]], [[129, 150], [287, 151], [287, 165], [129, 164]], [[36, 145], [73, 149], [72, 163], [35, 159]], [[527, 146], [552, 146], [552, 155], [527, 155]], [[102, 145], [120, 145], [120, 153], [102, 153]], [[371, 129], [503, 128], [503, 139], [371, 140]], [[99, 126], [193, 126], [193, 139], [99, 139]], [[322, 127], [337, 127], [337, 135], [322, 135]], [[37, 123], [77, 123], [77, 134], [37, 134]], [[324, 106], [337, 106], [337, 115], [324, 115]], [[309, 107], [315, 107], [315, 112], [309, 112]], [[372, 103], [501, 103], [501, 116], [372, 116]], [[349, 105], [360, 105], [360, 114], [349, 114]], [[38, 103], [80, 103], [80, 113], [38, 113]], [[99, 100], [205, 101], [205, 115], [99, 114]], [[306, 90], [317, 90], [317, 97], [306, 97]], [[347, 88], [362, 88], [362, 96], [347, 96]], [[321, 87], [340, 87], [340, 99], [321, 99]], [[358, 84], [513, 82], [513, 95], [358, 97]], [[41, 83], [89, 83], [89, 93], [41, 93]], [[94, 79], [241, 80], [241, 94], [94, 93]], [[313, 66], [394, 66], [394, 79], [313, 79]], [[242, 66], [288, 66], [288, 77], [242, 77]], [[185, 54], [220, 54], [220, 65], [185, 65]], [[469, 48], [547, 48], [547, 61], [469, 61]], [[423, 36], [436, 36], [436, 54], [423, 54]], [[465, 30], [551, 30], [551, 43], [465, 43]], [[207, 21], [329, 23], [328, 41], [207, 39]]]
00208502.jpg [[[247, 566], [282, 566], [282, 573], [247, 573]], [[558, 534], [629, 539], [627, 570], [556, 565]], [[205, 540], [284, 540], [284, 552], [205, 552]], [[143, 513], [189, 513], [189, 525], [143, 525]], [[249, 512], [307, 512], [307, 524], [249, 524]], [[44, 500], [118, 500], [118, 519], [44, 519]], [[467, 491], [556, 491], [556, 508], [467, 508]], [[667, 490], [678, 494], [675, 503], [664, 499]], [[788, 489], [794, 495], [789, 499], [783, 494]], [[726, 491], [737, 491], [737, 501], [726, 501]], [[42, 452], [117, 450], [117, 469], [42, 470]], [[175, 450], [236, 450], [236, 464], [175, 464]], [[614, 407], [638, 407], [638, 422], [614, 422]], [[95, 405], [119, 405], [119, 422], [95, 422]], [[49, 399], [64, 414], [50, 427], [36, 413]], [[209, 401], [226, 401], [226, 415], [209, 415]], [[40, 357], [58, 357], [58, 374], [40, 374]], [[94, 356], [119, 356], [119, 373], [94, 373]], [[188, 341], [246, 339], [247, 361], [189, 364]], [[459, 321], [549, 319], [549, 337], [460, 339]], [[459, 273], [551, 273], [551, 290], [459, 290]], [[563, 272], [735, 269], [735, 286], [564, 289]], [[517, 225], [547, 225], [547, 245], [517, 245]], [[459, 226], [480, 226], [480, 244], [459, 244]], [[621, 187], [673, 187], [673, 201], [621, 201]], [[457, 132], [548, 130], [548, 147], [458, 149]], [[572, 106], [787, 99], [787, 120], [573, 126]], [[122, 48], [290, 48], [290, 97], [122, 97]], [[539, 39], [708, 39], [708, 89], [539, 89]]]
00224225.jpg [[[134, 429], [153, 426], [157, 445], [138, 448]], [[202, 404], [478, 411], [476, 459], [201, 452]], [[205, 230], [469, 230], [469, 390], [205, 390]], [[131, 265], [172, 265], [172, 279], [131, 279]], [[345, 207], [456, 207], [456, 231], [345, 231]], [[199, 189], [346, 196], [344, 239], [197, 232]], [[10, 44], [157, 41], [158, 112], [11, 115]]]
00227746.jpg [[[190, 232], [258, 232], [258, 238], [190, 238]], [[160, 232], [183, 232], [183, 238], [160, 238]], [[123, 232], [150, 232], [150, 238], [123, 238]], [[290, 208], [345, 206], [346, 222], [291, 224]], [[172, 181], [249, 181], [249, 194], [172, 194]], [[143, 178], [165, 180], [162, 208], [140, 206]], [[142, 164], [157, 162], [160, 177], [145, 180]], [[173, 157], [203, 157], [203, 164], [173, 164]], [[200, 154], [347, 154], [347, 167], [200, 167]], [[144, 111], [277, 114], [277, 134], [144, 131]], [[201, 52], [387, 53], [386, 69], [201, 68]], [[141, 46], [192, 46], [192, 63], [141, 63]], [[40, 26], [61, 26], [61, 42], [40, 42]]]
00229605.jpg [[[743, 529], [881, 529], [881, 544], [743, 544]], [[236, 499], [589, 498], [589, 522], [236, 523]], [[6, 498], [227, 498], [227, 522], [6, 522]], [[735, 496], [883, 499], [883, 520], [734, 517]], [[606, 495], [716, 489], [718, 515], [608, 521]], [[4, 245], [863, 230], [864, 288], [5, 303]], [[478, 28], [883, 28], [883, 76], [478, 76]]]
00233011.jpg [[[63, 227], [291, 227], [291, 242], [63, 242]], [[12, 219], [41, 219], [41, 250], [12, 250]], [[61, 177], [119, 177], [119, 195], [61, 195]], [[11, 173], [40, 169], [44, 200], [14, 203]], [[61, 129], [147, 131], [147, 147], [61, 144]], [[12, 124], [43, 124], [43, 154], [12, 154]], [[125, 89], [238, 89], [238, 103], [125, 103]], [[148, 51], [216, 51], [216, 65], [148, 65]], [[258, 46], [353, 50], [352, 69], [257, 65]], [[9, 49], [52, 49], [52, 68], [9, 68]], [[277, 12], [345, 12], [345, 31], [277, 31]], [[28, 11], [74, 11], [74, 31], [28, 31]]]
00233625.jpg [[[375, 397], [632, 399], [632, 443], [375, 440]], [[71, 214], [932, 207], [933, 321], [71, 328]]]
00233634.jpg [[[215, 639], [261, 639], [261, 703], [215, 703]], [[523, 635], [570, 635], [570, 695], [523, 695]], [[643, 523], [682, 523], [682, 568], [643, 568]], [[97, 516], [152, 516], [152, 589], [97, 589]], [[755, 395], [760, 395], [760, 401], [755, 401]], [[26, 395], [32, 395], [32, 400], [26, 400]], [[678, 364], [728, 362], [731, 430], [681, 432]], [[54, 361], [107, 361], [107, 434], [54, 434]], [[78, 208], [155, 208], [155, 280], [78, 280]], [[643, 205], [693, 205], [693, 272], [643, 272]], [[210, 88], [260, 86], [263, 164], [213, 166]], [[363, 48], [426, 45], [430, 115], [367, 118]]]
00234400.jpg [[[446, 421], [738, 421], [738, 438], [446, 438]], [[157, 421], [454, 421], [454, 438], [157, 438]], [[158, 394], [652, 394], [652, 411], [158, 411]], [[40, 391], [127, 391], [127, 412], [40, 412]], [[158, 342], [304, 345], [304, 363], [158, 360]], [[38, 344], [123, 344], [123, 362], [38, 362]], [[520, 295], [703, 295], [703, 314], [520, 314]], [[394, 292], [483, 290], [484, 314], [394, 317]], [[157, 293], [270, 293], [270, 313], [157, 313]], [[37, 293], [125, 293], [125, 313], [37, 313]], [[156, 243], [358, 243], [358, 267], [156, 267]], [[36, 243], [82, 243], [82, 269], [36, 269]], [[29, 152], [158, 152], [158, 175], [29, 175]], [[282, 98], [507, 98], [507, 111], [282, 111]], [[315, 46], [475, 50], [474, 88], [314, 85]], [[518, 51], [663, 53], [662, 67], [517, 65]], [[487, 19], [706, 17], [706, 43], [487, 45]]]
00234883.jpg [[[344, 145], [355, 145], [355, 153], [344, 153]], [[66, 125], [316, 120], [317, 190], [67, 195]], [[79, 138], [109, 141], [108, 152], [78, 148]], [[72, 120], [120, 120], [120, 130], [72, 130]], [[383, 63], [504, 62], [504, 74], [383, 75]], [[58, 29], [365, 26], [366, 112], [59, 115]], [[387, 28], [501, 26], [501, 45], [387, 47]]]
test_add_0.jpg [[[311, 521], [391, 521], [391, 534], [311, 534]], [[277, 500], [424, 500], [424, 514], [277, 514]], [[261, 446], [437, 446], [437, 459], [261, 459]], [[212, 428], [485, 428], [485, 441], [212, 441]], [[247, 388], [457, 388], [457, 409], [247, 409]], [[222, 328], [474, 328], [474, 372], [222, 372]], [[208, 207], [492, 211], [490, 277], [207, 272]], [[266, 164], [422, 166], [421, 197], [265, 195]], [[18, 20], [201, 18], [201, 43], [18, 45]]]
test_add_1.png []
test_add_10.png [[[157, 124], [186, 124], [186, 172], [157, 172]], [[65, 117], [95, 117], [95, 168], [65, 168]], [[161, 106], [183, 106], [183, 127], [161, 127]], [[69, 100], [94, 100], [94, 128], [69, 128]], [[117, 46], [154, 45], [157, 174], [121, 175]], [[66, 34], [97, 34], [97, 112], [66, 112]]]
test_add_11.jpg [[[1525, 773], [1564, 756], [1575, 780], [1536, 798]], [[1390, 757], [1483, 757], [1483, 791], [1390, 791]], [[1013, 754], [1207, 754], [1207, 800], [1013, 800]], [[685, 755], [875, 755], [875, 796], [685, 796]], [[356, 753], [566, 747], [567, 793], [358, 798]], [[78, 751], [264, 745], [265, 793], [79, 798]], [[602, 647], [1152, 647], [1152, 703], [602, 703]], [[601, 564], [1148, 555], [1149, 611], [602, 620]], [[598, 480], [1066, 472], [1067, 526], [599, 535]], [[598, 393], [1090, 388], [1090, 439], [599, 444]], [[603, 306], [1057, 306], [1057, 357], [603, 357]], [[357, 184], [1517, 184], [1517, 261], [357, 261]], [[60, 43], [257, 37], [259, 83], [61, 89]], [[1305, 41], [1492, 41], [1492, 87], [1305, 87]], [[973, 40], [1171, 34], [1172, 80], [974, 86]], [[670, 40], [862, 34], [864, 80], [671, 86]], [[363, 34], [558, 34], [558, 85], [363, 85]]]
test_add_12.jpg [[[11, 592], [136, 594], [135, 613], [11, 611]], [[109, 521], [907, 526], [907, 569], [109, 565]], [[635, 451], [902, 448], [903, 478], [635, 481]], [[112, 447], [466, 449], [466, 486], [112, 483]], [[582, 306], [680, 304], [681, 348], [583, 351]], [[369, 261], [565, 266], [563, 357], [367, 353]], [[64, 85], [853, 88], [853, 161], [64, 159]]]
test_add_13.jpg [[[68, 94], [117, 97], [116, 114], [67, 111]]]
test_add_14.jpg [[[30, 97], [235, 95], [236, 127], [31, 129]], [[30, 52], [239, 50], [239, 86], [30, 87]]]
test_add_15.jpg [[[141, 253], [353, 253], [353, 266], [141, 266]], [[205, 214], [406, 219], [406, 232], [204, 227]], [[106, 212], [193, 213], [193, 227], [106, 226]], [[154, 156], [286, 161], [286, 174], [154, 170]], [[148, 136], [305, 142], [305, 156], [147, 150]], [[108, 137], [144, 137], [144, 148], [108, 148]], [[108, 102], [275, 109], [275, 125], [107, 117]], [[107, 72], [245, 79], [245, 96], [106, 88]], [[107, 39], [209, 42], [209, 62], [106, 59]]]
test_add_16.jpg [[[398, 842], [408, 842], [408, 852], [398, 852]], [[382, 742], [746, 742], [746, 776], [382, 776]], [[362, 703], [468, 703], [468, 725], [362, 725]], [[1552, 701], [1576, 701], [1576, 746], [1552, 746]], [[1256, 695], [1442, 695], [1442, 721], [1256, 721]], [[1244, 661], [1448, 661], [1448, 687], [1244, 687]], [[386, 645], [668, 645], [668, 679], [386, 679]], [[1228, 625], [1470, 623], [1470, 651], [1228, 653]], [[360, 604], [580, 604], [580, 629], [360, 629]], [[1202, 592], [1494, 592], [1494, 617], [1202, 617]], [[1166, 556], [1530, 556], [1530, 582], [1166, 582]], [[380, 552], [638, 552], [638, 586], [380, 586]], [[356, 502], [516, 502], [516, 536], [356, 536]], [[774, 260], [1124, 260], [1124, 300], [774, 300]], [[374, 210], [504, 210], [504, 300], [374, 300]], [[776, 212], [1088, 217], [1088, 252], [776, 248]]]
test_add_17.jpg [[[321, 255], [393, 258], [392, 271], [320, 269]], [[307, 222], [411, 228], [411, 241], [306, 236]], [[96, 136], [385, 143], [384, 208], [94, 201]], [[72, 95], [399, 103], [398, 124], [71, 117]], [[68, 76], [224, 79], [223, 93], [67, 90]], [[66, 59], [226, 62], [225, 76], [65, 74]]]
test_add_18.jpg [[[466, 788], [715, 790], [715, 813], [466, 811]], [[553, 752], [665, 757], [663, 791], [552, 786]], [[119, 539], [189, 539], [189, 570], [119, 570]], [[116, 473], [674, 486], [673, 528], [115, 516]], [[121, 429], [669, 441], [668, 470], [121, 457]], [[121, 375], [674, 381], [674, 410], [121, 404]], [[556, 262], [675, 264], [675, 278], [556, 277]], [[164, 259], [334, 259], [334, 273], [164, 273]], [[344, 195], [456, 197], [455, 220], [343, 217]], [[309, 175], [490, 175], [490, 190], [309, 190]], [[255, 128], [537, 131], [537, 169], [254, 165]], [[347, 92], [486, 94], [486, 109], [346, 107]], [[285, 41], [567, 49], [566, 82], [284, 74]], [[236, 32], [266, 32], [266, 60], [236, 60]]]
test_add_19.jpg [[[24, 294], [42, 294], [42, 302], [24, 302]], [[64, 293], [105, 293], [105, 303], [64, 303]], [[145, 287], [163, 287], [163, 304], [145, 304]], [[63, 280], [106, 280], [106, 290], [63, 290]], [[9, 281], [26, 281], [26, 288], [9, 288]], [[220, 279], [245, 279], [245, 291], [220, 291]], [[177, 279], [208, 279], [208, 290], [177, 290]], [[23, 279], [51, 279], [51, 290], [23, 290]], [[145, 278], [162, 278], [162, 292], [145, 292]], [[8, 267], [18, 267], [18, 276], [8, 276]], [[221, 265], [243, 265], [243, 277], [221, 277]], [[24, 265], [47, 265], [47, 277], [24, 277]], [[142, 263], [163, 263], [163, 279], [142, 279]], [[218, 252], [249, 252], [249, 265], [218, 265]], [[65, 253], [131, 253], [131, 263], [65, 263]], [[24, 252], [43, 252], [43, 264], [24, 264]], [[8, 253], [18, 253], [18, 262], [8, 262]], [[8, 240], [17, 240], [17, 249], [8, 249]], [[63, 237], [114, 237], [114, 251], [63, 251]], [[25, 236], [47, 239], [45, 251], [23, 249]], [[144, 234], [166, 237], [163, 253], [142, 249]], [[494, 226], [531, 226], [531, 239], [494, 239]], [[335, 226], [354, 226], [354, 237], [335, 237]], [[288, 226], [314, 226], [314, 237], [288, 237]], [[63, 226], [113, 226], [113, 236], [63, 236]], [[7, 227], [17, 227], [17, 234], [7, 234]], [[221, 225], [248, 225], [248, 235], [221, 235]], [[143, 225], [165, 222], [167, 234], [145, 237]], [[24, 224], [48, 224], [48, 238], [24, 238]], [[495, 213], [524, 213], [524, 224], [495, 224]], [[420, 212], [437, 212], [437, 225], [420, 225]], [[336, 212], [398, 212], [398, 223], [336, 223]], [[292, 212], [320, 212], [320, 223], [292, 223]], [[222, 212], [249, 212], [249, 223], [222, 223]], [[145, 212], [166, 212], [166, 223], [145, 223]], [[61, 211], [113, 209], [114, 222], [62, 224]], [[26, 211], [48, 211], [48, 223], [26, 223]], [[337, 199], [383, 199], [383, 209], [337, 209]], [[65, 200], [87, 200], [87, 207], [65, 207]], [[493, 197], [541, 197], [541, 211], [493, 211]], [[445, 202], [455, 196], [462, 206], [452, 212]], [[178, 198], [205, 198], [205, 208], [178, 208]], [[146, 199], [157, 199], [157, 208], [146, 208]], [[32, 194], [43, 204], [33, 214], [22, 203]], [[422, 193], [440, 201], [432, 215], [415, 207]], [[65, 186], [132, 186], [132, 196], [65, 196]], [[337, 185], [399, 185], [399, 196], [337, 196]], [[445, 190], [456, 182], [465, 191], [454, 200]], [[292, 188], [308, 182], [313, 193], [297, 200]], [[220, 183], [255, 183], [255, 197], [220, 197]], [[142, 184], [158, 184], [158, 197], [142, 197]], [[493, 182], [518, 182], [518, 197], [493, 197]], [[425, 180], [437, 191], [427, 202], [414, 190]], [[32, 179], [42, 189], [32, 199], [22, 189]], [[182, 179], [195, 185], [188, 198], [175, 192]], [[335, 172], [400, 169], [400, 183], [336, 185]], [[492, 170], [519, 170], [519, 185], [492, 185]], [[412, 177], [428, 164], [440, 178], [425, 190]], [[293, 171], [315, 171], [315, 185], [293, 185]], [[220, 170], [251, 170], [251, 184], [220, 184]], [[178, 172], [188, 172], [188, 183], [178, 183]], [[64, 172], [125, 170], [125, 181], [64, 182]], [[454, 168], [464, 176], [454, 185], [445, 176]], [[142, 172], [159, 168], [163, 180], [145, 185]], [[30, 165], [43, 174], [34, 186], [20, 177]], [[493, 160], [523, 160], [523, 170], [493, 170]], [[402, 161], [435, 161], [435, 168], [402, 168]], [[335, 159], [401, 159], [401, 169], [335, 169]], [[296, 159], [325, 159], [325, 170], [296, 170]], [[221, 158], [251, 158], [251, 169], [221, 169]], [[174, 161], [183, 156], [190, 167], [181, 172]], [[145, 158], [162, 158], [162, 170], [145, 170]], [[61, 158], [125, 157], [125, 168], [62, 169]], [[20, 161], [33, 154], [40, 167], [28, 174]], [[492, 143], [542, 143], [542, 157], [492, 157]], [[450, 144], [479, 144], [479, 157], [450, 157]], [[335, 143], [439, 143], [439, 156], [335, 156]], [[294, 143], [327, 143], [327, 157], [294, 157]], [[220, 143], [253, 143], [253, 157], [220, 157]], [[178, 145], [187, 145], [187, 156], [178, 156]], [[63, 144], [104, 144], [104, 155], [63, 155]], [[144, 140], [164, 145], [160, 159], [141, 154]], [[31, 137], [44, 149], [31, 162], [17, 149]], [[286, 135], [291, 135], [291, 140], [286, 140]], [[177, 133], [193, 133], [193, 144], [177, 144]], [[336, 132], [388, 132], [388, 141], [336, 141]], [[492, 131], [525, 131], [525, 141], [492, 141]], [[450, 131], [477, 131], [477, 141], [450, 141]], [[292, 131], [321, 131], [321, 141], [292, 141]], [[218, 132], [255, 130], [256, 141], [219, 144]], [[63, 131], [95, 131], [95, 141], [63, 141]], [[417, 130], [437, 130], [437, 141], [417, 141]], [[145, 130], [159, 130], [159, 143], [145, 143]], [[30, 124], [43, 133], [32, 147], [19, 138]], [[493, 118], [535, 118], [535, 129], [493, 129]], [[336, 118], [388, 118], [388, 129], [336, 129]], [[218, 118], [255, 118], [255, 128], [218, 128]], [[451, 117], [478, 117], [478, 129], [451, 129]], [[418, 117], [438, 117], [438, 130], [418, 130]], [[177, 116], [209, 116], [209, 130], [177, 130]], [[145, 117], [162, 117], [162, 130], [145, 130]], [[62, 116], [88, 116], [88, 131], [62, 131]], [[19, 121], [33, 111], [43, 124], [29, 134]], [[491, 107], [523, 107], [523, 113], [491, 113]], [[449, 107], [477, 107], [477, 113], [449, 113]], [[420, 107], [436, 107], [436, 113], [420, 113]], [[295, 107], [319, 107], [319, 114], [295, 114]], [[220, 107], [242, 107], [242, 113], [220, 113]], [[176, 107], [203, 107], [203, 113], [176, 113]], [[145, 107], [161, 107], [161, 114], [145, 114]], [[334, 105], [372, 105], [372, 114], [334, 114]], [[63, 106], [86, 106], [86, 113], [63, 113]], [[483, 89], [522, 89], [522, 99], [483, 99]], [[331, 88], [380, 88], [380, 99], [331, 99]], [[276, 88], [325, 88], [325, 99], [276, 99]], [[214, 88], [246, 88], [246, 99], [214, 99]], [[411, 86], [474, 86], [474, 100], [411, 100]], [[6, 86], [102, 86], [102, 100], [6, 100]], [[415, 66], [461, 66], [461, 77], [415, 77]], [[288, 66], [333, 66], [333, 77], [288, 77]], [[157, 64], [206, 64], [206, 78], [157, 78]], [[416, 48], [523, 49], [523, 63], [415, 62]], [[288, 49], [375, 49], [375, 63], [288, 63]], [[159, 49], [269, 49], [269, 62], [159, 62]], [[24, 53], [36, 46], [45, 59], [33, 67]], [[416, 36], [481, 36], [481, 46], [416, 46]], [[25, 38], [39, 32], [46, 46], [33, 52]], [[157, 34], [205, 34], [205, 47], [157, 47]], [[412, 4], [527, 4], [527, 17], [412, 17]], [[146, 4], [345, 2], [345, 15], [146, 17]]]
test_add_20.jpg [[[31, 346], [605, 346], [605, 370], [31, 370]], [[217, 294], [510, 294], [510, 322], [217, 322]], [[473, 271], [525, 271], [525, 286], [473, 286]], [[220, 267], [287, 267], [287, 286], [220, 286]], [[219, 239], [484, 239], [484, 263], [219, 263]], [[221, 217], [303, 217], [303, 234], [221, 234]], [[402, 192], [417, 192], [417, 205], [402, 205]], [[222, 187], [341, 187], [341, 207], [222, 207]], [[221, 162], [287, 162], [287, 180], [221, 180]], [[375, 122], [475, 124], [475, 146], [375, 143]], [[222, 124], [356, 122], [356, 143], [222, 146]], [[218, 81], [352, 84], [352, 116], [218, 113]], [[440, 35], [605, 35], [605, 60], [440, 60]], [[72, 16], [398, 16], [398, 44], [72, 44]]]
test_add_3.jpg [[[169, 327], [337, 326], [337, 341], [169, 342]], [[170, 288], [307, 290], [307, 312], [170, 310]], [[171, 221], [323, 221], [323, 234], [171, 234]], [[340, 221], [449, 217], [449, 231], [341, 234]], [[169, 201], [372, 201], [372, 214], [169, 214]], [[170, 183], [418, 183], [418, 196], [170, 196]], [[170, 149], [416, 149], [416, 163], [170, 163]], [[171, 119], [418, 119], [418, 140], [171, 140]], [[326, 64], [478, 64], [478, 91], [326, 91]], [[173, 64], [306, 60], [306, 89], [174, 93]]]
test_add_4.png []
test_add_5.png [[[48, 164], [108, 164], [108, 174], [48, 174]], [[52, 121], [169, 121], [169, 134], [52, 134]], [[50, 102], [165, 102], [165, 118], [50, 118]], [[52, 83], [164, 83], [164, 100], [52, 100]], [[51, 68], [166, 68], [166, 84], [51, 84]], [[51, 50], [145, 47], [145, 64], [52, 67]]]
test_add_6.jpg [[[123, 223], [219, 227], [218, 251], [122, 247]], [[172, 172], [186, 186], [172, 200], [158, 186]]]
test_add_7.jpg [[[48, 938], [174, 936], [174, 962], [48, 964]], [[227, 873], [629, 876], [628, 953], [226, 949]], [[56, 745], [638, 745], [638, 790], [56, 790]], [[150, 674], [545, 678], [544, 721], [150, 718]], [[73, 504], [633, 504], [633, 601], [73, 601]], [[59, 270], [655, 279], [652, 441], [56, 432]], [[513, 193], [553, 193], [553, 223], [513, 223]], [[61, 175], [532, 175], [532, 239], [61, 239]], [[533, 178], [642, 178], [642, 236], [533, 236]]]
test_add_8.jpg [[[251, 586], [454, 580], [454, 606], [252, 613]], [[107, 533], [457, 527], [457, 560], [108, 566]], [[336, 494], [384, 494], [384, 507], [336, 507]], [[27, 307], [355, 297], [356, 320], [28, 330]], [[22, 259], [445, 251], [445, 274], [23, 282]], [[78, 209], [445, 205], [445, 225], [78, 229]], [[160, 23], [319, 30], [317, 79], [158, 72]]]
test_add_9.png [[[266, 687], [486, 687], [486, 696], [266, 696]], [[196, 668], [554, 668], [554, 681], [196, 681]], [[154, 596], [597, 596], [597, 606], [154, 606]], [[215, 578], [541, 578], [541, 588], [215, 588]], [[134, 560], [615, 560], [615, 570], [134, 570]], [[85, 543], [665, 543], [665, 553], [85, 553]], [[96, 522], [653, 522], [653, 535], [96, 535]], [[362, 449], [389, 449], [389, 460], [362, 460]], [[238, 376], [513, 376], [513, 389], [238, 389]], [[177, 356], [574, 356], [574, 368], [177, 368]], [[344, 281], [408, 283], [407, 297], [343, 294]], [[257, 205], [493, 205], [493, 219], [257, 219]]]
tools/eval.py
浏览文件 @ f57a6d1b
......@@ -27,7 +27,7 @@ from ppocr.data import build_dataloader
from ppocr.modeling.architectures import build_model
from ppocr.postprocess import build_post_process
from ppocr.metrics import build_metric
from ppocr.utils.save_load import init_model, load_pretrained_params
from ppocr.utils.save_load import init_model, load_dygraph_params
from ppocr.utils.utility import print_dict
import tools.program as program
......@@ -61,7 +61,7 @@ def main():
else:
model_type = None
best_model_dict = init_model(config, model)
best_model_dict = load_dygraph_params(config, model, logger, None)
if len(best_model_dict):
logger.info('metric in ckpt ***************')
for k, v in best_model_dict.items():
......
tools/export_model.py
浏览文件 @ f57a6d1b
......@@ -93,6 +93,9 @@ def main():
for key in config["Architecture"]["Models"]:
config["Architecture"]["Models"][key]["Head"][
"out_channels"] = char_num
# just one final tensor needs to to exported for inference
config["Architecture"]["Models"][key][
"return_all_feats"] = False
else: # base rec model
config["Architecture"]["Head"]["out_channels"] = char_num
model = build_model(config["Architecture"])
......
tools/infer/predict_det.py
浏览文件 @ f57a6d1b
......@@ -30,7 +30,7 @@ from ppocr.utils.logging import get_logger
from ppocr.utils.utility import get_image_file_list, check_and_read_gif
from ppocr.data import create_operators, transform
from ppocr.postprocess import build_post_process
import json
logger = get_logger()
......@@ -243,6 +243,7 @@ if __name__ == "__main__":
if not os.path.exists(draw_img_save):
os.makedirs(draw_img_save)
save_results = []
for image_file in image_file_list:
img, flag = check_and_read_gif(image_file)
if not flag:
......@@ -256,8 +257,11 @@ if __name__ == "__main__":
if count > 0:
total_time += elapse
count += 1
logger.info("Predict time of {}: {}".format(image_file, elapse))
save_pred = os.path.basename(image_file) + "\t" + str(
json.dumps(np.array(dt_boxes).astype(np.int32).tolist())) + "\n"
save_results.append(save_pred)
logger.info(save_pred)
logger.info("The predict time of {}: {}".format(image_file, elapse))
src_im = utility.draw_text_det_res(dt_boxes, image_file)
img_name_pure = os.path.split(image_file)[-1]
img_path = os.path.join(draw_img_save,
......@@ -265,5 +269,8 @@ if __name__ == "__main__":
cv2.imwrite(img_path, src_im)
logger.info("The visualized image saved in {}".format(img_path))
with open(os.path.join(draw_img_save, "det_results.txt"), 'w') as f:
f.writelines(save_results)
f.close()
if args.benchmark:
text_detector.autolog.report()
tools/infer/utility.py
浏览文件 @ f57a6d1b
......@@ -35,7 +35,7 @@ def init_args():
parser.add_argument("--use_gpu", type=str2bool, default=True)
parser.add_argument("--ir_optim", type=str2bool, default=True)
parser.add_argument("--use_tensorrt", type=str2bool, default=False)
parser.add_argument("--min_subgraph_size", type=int, default=10)
parser.add_argument("--min_subgraph_size", type=int, default=15)
parser.add_argument("--precision", type=str, default="fp32")
parser.add_argument("--gpu_mem", type=int, default=500)
......
tools/infer_det.py
浏览文件 @ f57a6d1b
......@@ -34,23 +34,21 @@ import paddle
from ppocr.data import create_operators, transform
from ppocr.modeling.architectures import build_model
from ppocr.postprocess import build_post_process
from ppocr.utils.save_load import init_model
from ppocr.utils.save_load import init_model, load_dygraph_params
from ppocr.utils.utility import get_image_file_list
import tools.program as program
def draw_det_res(dt_boxes, config, img, img_name):
def draw_det_res(dt_boxes, config, img, img_name, save_path):
if len(dt_boxes) > 0:
import cv2
src_im = img
for box in dt_boxes:
box = box.astype(np.int32).reshape((-1, 1, 2))
cv2.polylines(src_im, [box], True, color=(255, 255, 0), thickness=2)
save_det_path = os.path.dirname(config['Global'][
'save_res_path']) + "/det_results/"
if not os.path.exists(save_det_path):
os.makedirs(save_det_path)
save_path = os.path.join(save_det_path, os.path.basename(img_name))
if not os.path.exists(save_path):
os.makedirs(save_path)
save_path = os.path.join(save_path, os.path.basename(img_name))
cv2.imwrite(save_path, src_im)
logger.info("The detected Image saved in {}".format(save_path))
......@@ -61,8 +59,7 @@ def main():
# build model
model = build_model(config['Architecture'])
init_model(config, model)
_ = load_dygraph_params(config, model, logger, None)
# build post process
post_process_class = build_post_process(config['PostProcess'])
......@@ -96,17 +93,41 @@ def main():
images = paddle.to_tensor(images)
preds = model(images)
post_result = post_process_class(preds, shape_list)
boxes = post_result[0]['points']
# write result
src_img = cv2.imread(file)
dt_boxes_json = []
for box in boxes:
tmp_json = {"transcription": ""}
tmp_json['points'] = box.tolist()
dt_boxes_json.append(tmp_json)
# parser boxes if post_result is dict
if isinstance(post_result, dict):
det_box_json = {}
for k in post_result.keys():
boxes = post_result[k][0]['points']
dt_boxes_list = []
for box in boxes:
tmp_json = {"transcription": ""}
tmp_json['points'] = box.tolist()
dt_boxes_list.append(tmp_json)
det_box_json[k] = dt_boxes_list
save_det_path = os.path.dirname(config['Global'][
'save_res_path']) + "/det_results_{}/".format(k)
draw_det_res(boxes, config, src_img, file, save_det_path)
else:
boxes = post_result[0]['points']
dt_boxes_json = []
# write result
for box in boxes:
tmp_json = {"transcription": ""}
tmp_json['points'] = box.tolist()
dt_boxes_json.append(tmp_json)
save_det_path = os.path.dirname(config['Global'][
'save_res_path']) + "/det_results/"
draw_det_res(boxes, config, src_img, file, save_det_path)
otstr = file + "\t" + json.dumps(dt_boxes_json) + "\n"
fout.write(otstr.encode())
src_img = cv2.imread(file)
draw_det_res(boxes, config, src_img, file)
save_det_path = os.path.dirname(config['Global'][
'save_res_path']) + "/det_results/"
draw_det_res(boxes, config, src_img, file, save_det_path)
logger.info("success!")
......
tools/infer_rec.py
浏览文件 @ f57a6d1b
......@@ -130,7 +130,7 @@ def main():
if len(post_result[key][0]) >= 2:
rec_info[key] = {
"label": post_result[key][0][0],
"score": post_result[key][0][1],
"score": float(post_result[key][0][1]),
}
info = json.dumps(rec_info)
else:
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或