update code_doc (#7667)

* update code_doc

* update code_doc

ppstructure/layout/README.md

@@ -23,7 +23,7 @@ English | [简体中文](README_ch.md)
 
 ## 1. Introduction
 
-Layout analysis refers to the regional division of documents in the form of pictures and the positioning of key areas, such as text, title, table, picture, etc. The layout analysis algorithm is based on the lightweight model PP-picodet of [PaddleDetection]( https://github.com/PaddlePaddle/PaddleDetection )
+Layout analysis refers to the regional division of documents in the form of pictures and the positioning of key areas, such as text, title, table, picture, etc. The layout analysis algorithm is based on the lightweight model PP-picodet of [PaddleDetection]( https://github.com/PaddlePaddle/PaddleDetection ), including English layout analysis, Chinese layout analysis and table layout analysis models.  English layout analysis models can detect document layout elements such as text, title, table, figure, list. Chinese layout analysis models can detect document layout elements such as text, figure, figure caption, table, table caption, header, footer, reference, and equation. Table layout analysis models can detect table regions.
 
 <div align="center">
     <img src="../docs/layout/layout.png" width="800">
@@ -152,7 +152,7 @@ We provide CDLA(Chinese layout analysis), TableBank(Table layout analysis)etc. d
 | [cTDaR2019_cTDaR](https://cndplab-founder.github.io/cTDaR2019/) | For form detection (TRACKA) and form identification (TRACKB).Image types include historical data sets (beginning with cTDaR_t0, such as CTDAR_T00872.jpg) and modern data sets (beginning with cTDaR_t1, CTDAR_T10482.jpg). |
 | [IIIT-AR-13K](http://cvit.iiit.ac.in/usodi/iiitar13k.php)    | Data sets constructed by manually annotating figures or pages from publicly available annual reports, containing 5 categories:table, figure, natural image, logo, and signature. |
 | [TableBank](https://github.com/doc-analysis/TableBank)       | For table detection and recognition of large datasets, including Word and Latex document formats |
-| [CDLA](https://github.com/buptlihang/CDLA)                   | Chinese document layout analysis data set, for Chinese literature (paper) scenarios, including 10 categories:Table, Figure, Figure caption, Table, Table caption, Header, Footer, Reference, Equation |
+| [CDLA](https://github.com/buptlihang/CDLA)                   | Chinese document layout analysis data set, for Chinese literature (paper) scenarios, including 10 categories:Text, Title, Figure, Figure caption, Table, Table caption, Header, Footer, Reference, Equation |
 | [DocBank](https://github.com/doc-analysis/DocBank)           | Large-scale dataset (500K document pages) constructed using weakly supervised methods for document layout analysis, containing 12 categories:Author, Caption, Date, Equation, Figure, Footer, List, Paragraph, Reference, Section, Table, Title |
 
 
@@ -175,7 +175,7 @@ If the test image is Chinese, the pre-trained model of Chinese CDLA dataset can
 
 ### 5.1. Train
 
-Train:
+Start training with the PaddleDetection [layout analysis profile](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.5/configs/picodet/legacy_model/application/layout_analysis)
 
 * Modify Profile
 

ppstructure/layout/README_ch.md

@@ -22,7 +22,7 @@
 
 ## 1. 简介
 
-版面分析指的是对图片形式的文档进行区域划分，定位其中的关键区域，如文字、标题、表格、图片等。版面分析算法基于[PaddleDetection](https://github.com/PaddlePaddle/PaddleDetection)的轻量模型PP-PicoDet进行开发。
+版面分析指的是对图片形式的文档进行区域划分，定位其中的关键区域，如文字、标题、表格、图片等。版面分析算法基于[PaddleDetection](https://github.com/PaddlePaddle/PaddleDetection)的轻量模型PP-PicoDet进行开发，包含英文、中文、表格版面分析3类模型。其中，英文模型支持Text、Title、Tale、Figure、List5类区域的检测，中文模型支持Text、Title、Figure、Figure caption、Table、Table caption、Header、Footer、Reference、Equation10类区域的检测，表格版面分析支持Table区域的检测，版面分析效果如下图所示：
 
 <div align="center">
     <img src="../docs/layout/layout.png" width="800">
@@ -152,7 +152,7 @@ json文件包含所有图像的标注，数据以字典嵌套的方式存放，
 | ------------------------------------------------------------ | ------------------------------------------------------------ |
 | [cTDaR2019_cTDaR](https://cndplab-founder.github.io/cTDaR2019/) | 用于表格检测(TRACKA)和表格识别(TRACKB)。图片类型包含历史数据集(以cTDaR_t0开头，如cTDaR_t00872.jpg)和现代数据集(以cTDaR_t1开头，cTDaR_t10482.jpg)。 |
 | [IIIT-AR-13K](http://cvit.iiit.ac.in/usodi/iiitar13k.php)    | 手动注释公开的年度报告中的图形或页面而构建的数据集，包含5类：table, figure, natural image, logo, and signature |
-| [CDLA](https://github.com/buptlihang/CDLA)                   | 中文文档版面分析数据集，面向中文文献类（论文）场景，包含10类：Table、Figure、Figure caption、Table、Table caption、Header、Footer、Reference、Equation |
+| [CDLA](https://github.com/buptlihang/CDLA)                   | 中文文档版面分析数据集，面向中文文献类（论文）场景，包含10类：Text、Title、Figure、Figure caption、Table、Table caption、Header、Footer、Reference、Equation |
 | [TableBank](https://github.com/doc-analysis/TableBank)       | 用于表格检测和识别大型数据集，包含Word和Latex2种文档格式     |
 | [DocBank](https://github.com/doc-analysis/DocBank)           | 使用弱监督方法构建的大规模数据集(500K文档页面)，用于文档布局分析，包含12类：Author、Caption、Date、Equation、Figure、Footer、List、Paragraph、Reference、Section、Table、Title |
 
@@ -161,7 +161,7 @@ json文件包含所有图像的标注，数据以字典嵌套的方式存放，
 
 提供了训练脚本、评估脚本和预测脚本，本节将以PubLayNet预训练模型为例进行讲解。
 
-如果不希望训练，直接体验后面的模型评估、预测、动转静、推理的流程，可以下载提供的预训练模型(PubLayNet数据集)，并跳过本部分。
+如果不希望训练，直接体验后面的模型评估、预测、动转静、推理的流程，可以下载提供的预训练模型(PubLayNet数据集)，并跳过5.1和5.2。
 
 ```
 mkdir pretrained_model
@@ -176,7 +176,7 @@ wget https://paddleocr.bj.bcebos.com/ppstructure/models/layout/picodet_lcnet_x1_
 
 ### 5.1. 启动训练
 
-开始训练:
+使用PaddleDetection[版面分析配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.5/configs/picodet/legacy_model/application/layout_analysis)启动训练
 
 * 修改配置文件
 

ppstructure/predict_system.py

@@ -254,8 +254,7 @@ def main(args):
 
         if args.recovery and all_res != []:
             try:
-                convert_info_docx(img, all_res, save_folder, img_name,
-                                  args.save_pdf)
+                convert_info_docx(img, all_res, save_folder, img_name)
             except Exception as ex:
                 logger.error("error in layout recovery image:{}, err msg: {}".
                              format(image_file, ex))

ppstructure/recovery/README.md

@@ -82,8 +82,11 @@ Through layout analysis, we divided the image/PDF documents into regions, locate
 
 We can restore the test picture through the layout information, OCR detection and recognition structure, table information, and saved pictures.
 
-The whl package is also provided  for quick use, see [quickstart](../docs/quickstart_en.md) for details.
+The whl package is also provided  for quick use, follow the above code, for more infomation please refer to [quickstart](../docs/quickstart_en.md) for details.
 
+```bash
+paddleocr --image_dir=ppstructure/docs/table/1.png --type=structure --recovery=true --lang='en'
+```
 
 <a name="3.1"></a>
 ### 3.1 Download models

ppstructure/recovery/README_ch.md

@@ -83,7 +83,16 @@ python3 -m pip install -r ppstructure/recovery/requirements.txt
 
 我们通过版面信息、OCR检测和识别结构、表格信息、保存的图片，对测试图片进行恢复即可。
 
-提供如下代码实现版面恢复，也提供了whl包的形式方便快速使用，详见 [quickstart](../docs/quickstart.md)。
+提供如下代码实现版面恢复，也提供了whl包的形式方便快速使用，代码如下，更多信息详见 [quickstart](../docs/quickstart.md)。
+
+```bash
+# 中文测试图
+paddleocr --image_dir=ppstructure/docs/table/1.png --type=structure --recovery=true
+# 英文测试图
+paddleocr --image_dir=ppstructure/docs/table/1.png --type=structure --recovery=true --lang='en'
+# pdf测试文件
+paddleocr --image_dir=ppstructure/recovery/UnrealText.pdf --type=structure --recovery=true --lang='en'
+```
 
 <a name="3.1"></a>
 

ppstructure/recovery/recovery_to_doc.py

@@ -28,7 +28,7 @@ from ppocr.utils.logging import get_logger
 logger = get_logger()
 
 
-def convert_info_docx(img, res, save_folder, img_name, save_pdf=False):
+def convert_info_docx(img, res, save_folder, img_name):
     doc = Document()
     doc.styles['Normal'].font.name = 'Times New Roman'
     doc.styles['Normal']._element.rPr.rFonts.set(qn('w:eastAsia'), u'宋体')
@@ -60,14 +60,9 @@ def convert_info_docx(img, res, save_folder, img_name, save_pdf=False):
         elif region['type'].lower() == 'title':
             doc.add_heading(region['res'][0]['text'])
         elif region['type'].lower() == 'table':
-            paragraph = doc.add_paragraph()
-            new_parser = HtmlToDocx()
-            new_parser.table_style = 'TableGrid'
-            table = new_parser.handle_table(html=region['res']['html'])
-            new_table = deepcopy(table)
-            new_table.alignment = WD_TABLE_ALIGNMENT.CENTER
-            paragraph.add_run().element.addnext(new_table._tbl)
-
+            parser = HtmlToDocx()
+            parser.table_style = 'TableGrid'
+            parser.handle_table(region['res']['html'], doc)
         else:
             paragraph = doc.add_paragraph()
             paragraph_format = paragraph.paragraph_format
@@ -82,13 +77,6 @@ def convert_info_docx(img, res, save_folder, img_name, save_pdf=False):
     doc.save(docx_path)
     logger.info('docx save to {}'.format(docx_path))
 
-    # save to pdf
-    if save_pdf:
-        pdf_path = os.path.join(save_folder, '{}.pdf'.format(img_name))
-        from docx2pdf import convert
-        convert(docx_path, pdf_path)
-        logger.info('pdf save to {}'.format(pdf_path))
-
 
 def sorted_layout_boxes(res, w):
     """

ppstructure/recovery/requirements.txt

 python-docx
-docx2pdf
 PyMuPDF
 beautifulsoup4
\ No newline at end of file

ppstructure/utility.py

@@ -90,11 +90,6 @@ def init_args():
         type=str2bool,
         default=False,
         help='Whether to enable layout of recovery')
-    parser.add_argument(
-        "--save_pdf",
-        type=str2bool,
-        default=False,
-        help='Whether to save pdf file')
 
     return parser
 
@@ -108,7 +103,38 @@ def draw_structure_result(image, result, font_path):
     if isinstance(image, np.ndarray):
         image = Image.fromarray(image)
     boxes, txts, scores = [], [], []
+
+    img_layout = image.copy()
+    draw_layout = ImageDraw.Draw(img_layout)
+    text_color = (255, 255, 255)
+    text_background_color = (80, 127, 255)
+    catid2color = {}
+    font_size = 15
+    font = ImageFont.truetype(font_path, font_size, encoding="utf-8")
+
     for region in result:
+        if region['type'] not in catid2color:
+            box_color = (random.randint(0, 255), random.randint(0, 255),
+                         random.randint(0, 255))
+            catid2color[region['type']] = box_color
+        else:
+            box_color = catid2color[region['type']]
+        box_layout = region['bbox']
+        draw_layout.rectangle(
+            [(box_layout[0], box_layout[1]), (box_layout[2], box_layout[3])],
+            outline=box_color,
+            width=3)
+        text_w, text_h = font.getsize(region['type'])
+        draw_layout.rectangle(
+            [(box_layout[0], box_layout[1]),
+             (box_layout[0] + text_w, box_layout[1] + text_h)],
+            fill=text_background_color)
+        draw_layout.text(
+            (box_layout[0], box_layout[1]),
+            region['type'],
+            fill=text_color,
+            font=font)
+
         if region['type'] == 'table':
             pass
         else:
@@ -116,6 +142,7 @@ def draw_structure_result(image, result, font_path):
                 boxes.append(np.array(text_result['text_region']))
                 txts.append(text_result['text'])
                 scores.append(text_result['confidence'])
+
     im_show = draw_ocr_box_txt(
-        image, boxes, txts, scores, font_path=font_path, drop_score=0)
+        img_layout, boxes, txts, scores, font_path=font_path, drop_score=0)
     return im_show