diff --git a/doc/doc_ch/whl.md b/doc/doc_ch/whl.md index 957c6926b15fad3091265da9295f5ad820fe6a26..0c969988a649ba6cea0115abd1ba0aa5b19ea892 100644 --- a/doc/doc_ch/whl.md +++ b/doc/doc_ch/whl.md @@ -356,3 +356,4 @@ im_show.save('result.jpg') | rec | 前向时是否启动识别 | TRUE | | cls | 前向时是否启动分类 (命令行模式下使用use_angle_cls控制前向是否启动分类) | FALSE | | show_log | 是否打印det和rec等信息 | FALSE | +| type | 执行ocr或者表格结构化, 值可选['ocr','structure'] | ocr | diff --git a/doc/doc_en/whl_en.md b/doc/doc_en/whl_en.md index b9909f498e830309eaad952df9171cd63b6f5e7b..2d68845b18ef8937c626179c2e0a746201341bec 100644 --- a/doc/doc_en/whl_en.md +++ b/doc/doc_en/whl_en.md @@ -362,5 +362,5 @@ im_show.save('result.jpg') | det | Enable detction when `ppocr.ocr` func exec | TRUE | | rec | Enable recognition when `ppocr.ocr` func exec | TRUE | | cls | Enable classification when `ppocr.ocr` func exec((Use use_angle_cls in command line mode to control whether to start classification in the forward direction) | FALSE | -| show_log | Whether to print log in det and rec - | FALSE | \ No newline at end of file +| show_log | Whether to print log in det and rec | FALSE | +| type | Perform ocr or table structuring, the value is selected in ['ocr','structure'] | ocr | \ No newline at end of file diff --git a/paddleocr.py b/paddleocr.py index 80fe0b981795064d20e1a06b0dd78efcd86e5e8d..d14f7b8b098ce2da8f387bb526bb450432eb4ae9 100644 --- a/paddleocr.py +++ b/paddleocr.py @@ -33,7 +33,7 @@ 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'] +__all__ = ['PaddleOCR', 'PPStructure', 'draw_ocr', 'draw_structure_result', 'save_structure_res'] model_urls = { 'det': { @@ -153,6 +153,42 @@ def parse_args(mMain=True): return argparse.Namespace(**inference_args_dict) +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' + ] + 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' + ] + devanagari_lang = [ + 'hi', 'mr', 'ne', 'bh', 'mai', 'ang', 'bho', 'mah', 'sck', 'new', + 'gom', 'sa', 'bgc' + ] + if lang in latin_lang: + lang = "latin" + elif lang in arabic_lang: + lang = "arabic" + elif lang in cyrillic_lang: + lang = "cyrillic" + elif lang in devanagari_lang: + lang = "devanagari" + assert lang in model_urls[ + 'rec'], 'param lang must in {}, but got {}'.format( + model_urls['rec'].keys(), lang) + if lang == "ch": + det_lang = "ch" + elif lang == 'structure': + det_lang = 'structure' + else: + det_lang = "en" + return lang, det_lang + + class PaddleOCR(predict_system.TextSystem): def __init__(self, **kwargs): """ @@ -165,42 +201,7 @@ class PaddleOCR(predict_system.TextSystem): if not params.show_log: logger.setLevel(logging.INFO) self.use_angle_cls = params.use_angle_cls - lang = params.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' - ] - 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' - ] - devanagari_lang = [ - 'hi', 'mr', 'ne', 'bh', 'mai', 'ang', 'bho', 'mah', 'sck', 'new', - 'gom', 'sa', 'bgc' - ] - if lang in latin_lang: - lang = "latin" - elif lang in arabic_lang: - lang = "arabic" - elif lang in cyrillic_lang: - lang = "cyrillic" - elif lang in devanagari_lang: - lang = "devanagari" - assert lang in model_urls[ - 'rec'], 'param lang must in {}, but got {}'.format( - model_urls['rec'].keys(), lang) - if lang == "ch": - det_lang = "ch" - else: - det_lang = "en" - use_inner_dict = False - if params.rec_char_dict_path is None: - use_inner_dict = True - params.rec_char_dict_path = model_urls['rec'][lang][ - 'dict_path'] + lang, det_lang = parse_lang(params.lang) # init model dir params.det_model_dir, det_url = confirm_model_dir_url(params.det_model_dir, @@ -223,9 +224,9 @@ class PaddleOCR(predict_system.TextSystem): if params.rec_algorithm not in SUPPORT_REC_MODEL: logger.error('rec_algorithm must in {}'.format(SUPPORT_REC_MODEL)) sys.exit(0) - if use_inner_dict: - params.rec_char_dict_path = str( - Path(__file__).parent / params.rec_char_dict_path) + + if params.rec_char_dict_path is None: + params.rec_char_dict_path = str(Path(__file__).parent / model_urls['rec'][lang]['dict_path']) print(params) # init det_model and rec_model @@ -289,16 +290,17 @@ class PPStructure(OCRSystem): params.__dict__.update(**kwargs) if not params.show_log: logger.setLevel(logging.INFO) - params.use_angle_cls = False + 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, 'structure', 'det'), - model_urls['det']['structure']) + 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, 'structure', 'rec'), - model_urls['rec']['structure']['url']) + 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, 'structure', 'table'), + os.path.join(BASE_DIR, VERSION, 'ocr', 'table'), model_urls['table']['url']) # download model maybe_download(params.det_model_dir, det_url) @@ -306,16 +308,9 @@ class PPStructure(OCRSystem): maybe_download(params.table_model_dir, table_url) if params.rec_char_dict_path is None: - params.rec_char_type = 'EN' - if os.path.exists(str(Path(__file__).parent / model_urls['rec']['structure']['dict_path'])): - params.rec_char_dict_path = str(Path(__file__).parent / model_urls['rec']['structure']['dict_path']) - else: - params.rec_char_dict_path = str(Path(__file__).parent.parent / model_urls['rec']['structure']['dict_path']) + params.rec_char_dict_path = str(Path(__file__).parent / model_urls['rec'][lang]['dict_path']) if params.table_char_dict_path is None: - if os.path.exists(str(Path(__file__).parent / model_urls['table']['dict_path'])): - params.table_char_dict_path = str(Path(__file__).parent / model_urls['table']['dict_path']) - else: - params.table_char_dict_path = str(Path(__file__).parent.parent / model_urls['table']['dict_path']) + params.table_char_dict_path = str(Path(__file__).parent / model_urls['table']['dict_path']) print(params) super().__init__(params) @@ -354,9 +349,9 @@ def main(): if len(image_file_list) == 0: logger.error('no images find in {}'.format(args.image_dir)) return - if args.type=='ocr': + if args.type == 'ocr': engine = PaddleOCR(**(args.__dict__)) - elif args.type=='structure': + elif args.type == 'structure': engine = PPStructure(**(args.__dict__)) else: raise NotImplementedError @@ -366,9 +361,9 @@ def main(): logger.info('{}{}{}'.format('*' * 10, img_path, '*' * 10)) if args.type == 'ocr': result = engine.ocr(img_path, - det=args.det, - rec=args.rec, - cls=args.use_angle_cls) + det=args.det, + rec=args.rec, + cls=args.use_angle_cls) if result is not None: for line in result: logger.info(line) @@ -376,4 +371,4 @@ def main(): result = engine(img_path) for item in result: logger.info(item['res']) - save_structure_res(result, args.output, img_name) \ No newline at end of file + save_structure_res(result, args.output, img_name) diff --git a/ppstructure/README.md b/ppstructure/README.md index 90cd412df038a59ab6555b1ff632f99e2d32bb74..00e8ba8f2aa8f1c53a70e5c502852c0f028aaeef 100644 --- a/ppstructure/README.md +++ b/ppstructure/README.md @@ -1,26 +1,17 @@ -# PaddleStructure +# PPStructure -PaddleStructure is an OCR toolkit for complex layout analysis. It can divide document data in the form of pictures into **text, table, title, picture and list** 5 types of areas, and extract the table area as excel +PPStructure is an OCR toolkit for complex layout analysis. It can divide document data in the form of pictures into **text, table, title, picture and list** 5 types of areas, and extract the table area as excel ## 1. Quick start ### install -**install layoutparser** -```sh -pip3 install -U premailer paddleocr https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl -``` -**install paddlestructure** +**install paddleocr** -install by pypi +ref to [paddleocr whl doc](../doc/doc_en/whl_en.md) -```bash -pip install paddlestructure -``` - -build own whl package and install -```bash -python3 setup.py bdist_wheel -pip3 install dist/paddlestructure-x.x.x-py3-none-any.whl # x.x.x is the version of paddlestructure +**install layoutparser** +```sh +pip3 install -U premailer https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl ``` ### 1.2 Use @@ -28,7 +19,7 @@ pip3 install dist/paddlestructure-x.x.x-py3-none-any.whl # x.x.x is the version #### 1.2.1 Use by command line ```bash -paddlestructure --image_dir=../doc/table/1.png +paddleocr --image_dir=../doc/table/1.png --type=structure ``` #### 1.2.2 Use by code @@ -36,29 +27,29 @@ paddlestructure --image_dir=../doc/table/1.png ```python import os import cv2 -from paddlestructure import PaddleStructure,draw_result,save_res +from paddleocr import PPStructure,draw_structure_result,save_structure_res -table_engine = PaddleStructure(show_log=True) +table_engine = PPStructure(show_log=True) save_folder = './output/table' img_path = '../doc/table/1.png' img = cv2.imread(img_path) result = table_engine(img) -save_res(result, save_folder,os.path.basename(img_path).split('.')[0]) +save_structure_res(result, save_folder,os.path.basename(img_path).split('.')[0]) for line in result: print(line) from PIL import Image -font_path = '../doc/fonts/simfang.ttf' # PaddleOCR下提供字体包 +font_path = '../doc/fonts/simfang.ttf' image = Image.open(img_path).convert('RGB') -im_show = draw_result(image, result,font_path=font_path) +im_show = draw_structure_result(image, result,font_path=font_path) im_show = Image.fromarray(im_show) im_show.save('result.jpg') ``` #### 1.2.3 返回结果说明 -The return result of PaddleStructure is a list composed of a dict, an example is as follows +The return result of PPStructure is a list composed of a dict, an example is as follows ```shell [ @@ -91,12 +82,12 @@ 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 the excel file name will be the coordinates of the table in the image. -## 2. PaddleStructure Pipeline +## 2. PPStructure Pipeline the process is as follows ![pipeline](../doc/table/pipeline_en.jpg) -In PaddleStructure, 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. +In PPStructure, 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. ### 2.1 LayoutParser diff --git a/ppstructure/README_ch.md b/ppstructure/README_ch.md index 7ae55534309ab48caecf8de1ae20c0536b49823e..296b730e29f3b5a1b6029f02c5c9875b9127709b 100644 --- a/ppstructure/README_ch.md +++ b/ppstructure/README_ch.md @@ -1,4 +1,4 @@ -# PaddleStructure +# PPStructure PaddleStructure是一个用于复杂版面分析的OCR工具包，其能够对图片形式的文档数据划分**文字、表格、标题、图片以及列表**5类区域，并将表格区域提取为excel @@ -6,29 +6,21 @@ PaddleStructure是一个用于复杂版面分析的OCR工具包，其能够对 ### 1.1 安装 +**安装 paddleocr** + +参考 [paddleocr whl文档](../doc/doc_ch/whl.md) + **安装 layoutparser** ```sh pip3 install -U premailer paddleocr https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl ``` -**安装 paddlestructure** - -pip安装 -```bash -pip install paddlestructure -``` - -本地构建并安装 -```bash -python3 setup.py bdist_wheel -pip3 install dist/paddlestructure-x.x.x-py3-none-any.whl # x.x.x是 paddlestructure 的版本号 -``` -### 1.2 PaddleStructure whl包使用 +### 1.2 PPStructure whl包使用 #### 1.2.1 命令行使用 ```bash -paddlestructure --image_dir=../doc/table/1.png +paddleocr --image_dir=../doc/table/1.png --type=structure ``` #### 1.2.2 Python脚本使用 @@ -36,15 +28,15 @@ paddlestructure --image_dir=../doc/table/1.png ```python import os import cv2 -from paddlestructure import PaddleStructure,draw_result,save_res +from paddleocr import PPStructure,draw_structure_result,save_structure_res -table_engine = PaddleStructure(show_log=True) +table_engine = PPStructure(show_log=True) save_folder = './output/table' img_path = '../doc/table/1.png' img = cv2.imread(img_path) result = table_engine(img) -save_res(result, save_folder,os.path.basename(img_path).split('.')[0]) +save_structure_res(result, save_folder,os.path.basename(img_path).split('.')[0]) for line in result: print(line) @@ -53,7 +45,7 @@ from PIL import Image font_path = '../doc/fonts/simfang.ttf' # PaddleOCR下提供字体包 image = Image.open(img_path).convert('RGB') -im_show = draw_result(image, result,font_path=font_path) +im_show = draw_structure_result(image, result,font_path=font_path) im_show = Image.fromarray(im_show) im_show.save('result.jpg') ``` @@ -93,12 +85,12 @@ dict 里各个字段说明如下 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，excel文件名为表格在图片里的坐标。 -## 2. PaddleStructure Pipeline +## 2. PPStructure Pipeline 流程如下 ![pipeline](../doc/table/pipeline.jpg) -在PaddleStructure中，图片会先经由layoutparser进行版面分析，在版面分析中，会对图片里的区域进行分类，包括**文字、标题、图片、列表和表格**5类。对于前4类区域，直接使用PP-OCR完成对应区域文字检测与识别。对于表格类区域，经过Table OCR处理后，表格图片转换为相同表格样式的Excel文件。 +在PPStructure中，图片会先经由layoutparser进行版面分析，在版面分析中，会对图片里的区域进行分类，包括**文字、标题、图片、列表和表格**5类。对于前4类区域，直接使用PP-OCR完成对应区域文字检测与识别。对于表格类区域，经过Table OCR处理后，表格图片转换为相同表格样式的Excel文件。 ### 2.1 版面分析