Skip to content

GitLab

"Deepspeed/Megatron-LM/vscode:/vscode.git/clone" did not exist on "5394b117faed9282f1bf7b71685d41f9bbdcc7d1"
Commit dad8397f authored by LDOUBLEV's avatar LDOUBLEV
Browse files

Merge branch 'dygraph' of https://github.com/PaddlePaddle/PaddleOCR into test_v7

parents 5801311c da5d7ee3
Show whitespace changes
Inline Side-by-side
Showing with 283 additions and 373 deletions
test1/predict_system.py → ppstructure/predict_system.py
View file @ dad8397f
...@@ -26,26 +26,33 @@ import numpy as np ...@@ -26,26 +26,33 @@ import numpy as np
import time import time
import logging import logging
import layoutparser as lp
from ppocr.utils.utility import get_image_file_list, check_and_read_gif from ppocr.utils.utility import get_image_file_list, check_and_read_gif
from ppocr.utils.logging import get_logger from ppocr.utils.logging import get_logger
from tools.infer.predict_system import TextSystem from tools.infer.predict_system import TextSystem
from test1.table.predict_table import TableSystem, to_excel from ppstructure.table.predict_table import TableSystem, to_excel
from test1.utility import parse_args, draw_result from ppstructure.utility import parse_args, draw_structure_result
logger = get_logger() logger = get_logger()
class OCRSystem(object): class OCRSystem(object):
def __init__(self, args): def __init__(self, args):
args.det_limit_type = 'resize_long' import layoutparser as lp
# args.det_limit_type = 'resize_long'
args.drop_score = 0 args.drop_score = 0
if not args.show_log: if not args.show_log:
logger.setLevel(logging.INFO) logger.setLevel(logging.INFO)
self.text_system = TextSystem(args) self.text_system = TextSystem(args)
self.table_system = TableSystem(args, self.text_system.text_detector, self.text_system.text_recognizer) self.table_system = TableSystem(args, self.text_system.text_detector, self.text_system.text_recognizer)
self.table_layout = lp.PaddleDetectionLayoutModel("lp://PubLayNet/ppyolov2_r50vd_dcn_365e_publaynet/config",
config_path = None
model_path = None
if os.path.isdir(args.layout_path_model):
model_path = args.layout_path_model
else:
config_path = args.layout_path_model
self.table_layout = lp.PaddleDetectionLayoutModel(config_path=config_path,
model_path=model_path,
threshold=0.5, enable_mkldnn=args.enable_mkldnn, threshold=0.5, enable_mkldnn=args.enable_mkldnn,
enforce_cpu=not args.use_gpu, thread_num=args.cpu_threads) enforce_cpu=not args.use_gpu, thread_num=args.cpu_threads)
self.use_angle_cls = args.use_angle_cls self.use_angle_cls = args.use_angle_cls
...@@ -65,24 +72,35 @@ class OCRSystem(object): ...@@ -65,24 +72,35 @@ class OCRSystem(object):
filter_boxes, filter_rec_res = self.text_system(roi_img) filter_boxes, filter_rec_res = self.text_system(roi_img)
filter_boxes = [x + [x1, y1] for x in filter_boxes] filter_boxes = [x + [x1, y1] for x in filter_boxes]
filter_boxes = [x.reshape(-1).tolist() for x in filter_boxes] filter_boxes = [x.reshape(-1).tolist() for x in filter_boxes]
# remove style char
res = (filter_boxes, filter_rec_res) style_token = ['<strike>', '<strike>', '<sup>', '</sub>', '<b>', '</b>', '<sub>', '</sup>',
res_list.append({'type': region.type, 'bbox': [x1, y1, x2, y2], 'res': res}) '<overline>', '</overline>', '<underline>', '</underline>', '<i>', '</i>']
filter_rec_res_tmp = []
for rec_res in filter_rec_res:
rec_str, rec_conf = rec_res
for token in style_token:
if token in rec_str:
rec_str = rec_str.replace(token, '')
filter_rec_res_tmp.append((rec_str, rec_conf))
res = (filter_boxes, filter_rec_res_tmp)
res_list.append({'type': region.type, 'bbox': [x1, y1, x2, y2], 'img': roi_img, 'res': res})
return res_list return res_list
def save_res(res, save_folder, img_name): def save_structure_res(res, save_folder, img_name):
excel_save_folder = os.path.join(save_folder, img_name) excel_save_folder = os.path.join(save_folder, img_name)
os.makedirs(excel_save_folder, exist_ok=True) os.makedirs(excel_save_folder, exist_ok=True)
# save res # save res
with open(os.path.join(excel_save_folder, 'res.txt'), 'w', encoding='utf8') as f:
for region in res: for region in res:
if region['type'] == 'Table': if region['type'] == 'Table':
excel_path = os.path.join(excel_save_folder, '{}.xlsx'.format(region['bbox'])) excel_path = os.path.join(excel_save_folder, '{}.xlsx'.format(region['bbox']))
to_excel(region['res'], excel_path) to_excel(region['res'], excel_path)
elif region['type'] == 'Figure': if region['type'] == 'Figure':
pass roi_img = region['img']
img_path = os.path.join(excel_save_folder, '{}.jpg'.format(region['bbox']))
cv2.imwrite(img_path, roi_img)
else: else:
with open(os.path.join(excel_save_folder, 'res.txt'), 'a', encoding='utf8') as f:
for box, rec_res in zip(region['res'][0], region['res'][1]): for box, rec_res in zip(region['res'][0], region['res'][1]):
f.write('{}\t{}\n'.format(np.array(box).reshape(-1).tolist(), rec_res)) f.write('{}\t{}\n'.format(np.array(box).reshape(-1).tolist(), rec_res))
...@@ -108,8 +126,8 @@ def main(args): ...@@ -108,8 +126,8 @@ def main(args):
continue continue
starttime = time.time() starttime = time.time()
res = structure_sys(img) res = structure_sys(img)
save_res(res, save_folder, img_name) save_structure_res(res, save_folder, img_name)
draw_img = draw_result(img, res, args.vis_font_path) draw_img = draw_structure_result(img, res, args.vis_font_path)
cv2.imwrite(os.path.join(save_folder, img_name, 'show.jpg'), draw_img) cv2.imwrite(os.path.join(save_folder, img_name, 'show.jpg'), draw_img)
logger.info('result save to {}'.format(os.path.join(save_folder, img_name))) logger.info('result save to {}'.format(os.path.join(save_folder, img_name)))
elapse = time.time() - starttime elapse = time.time() - starttime
......
ppstructure/table/README.md 0 → 100644
View file @ dad8397f
# Table Recognition
## 1. pipeline
The table recognition mainly contains three models
1. Single line text detection-DB
2. Single line text recognition-CRNN
3. Table structure and cell coordinate prediction-RARE
The table recognition flow chart is as follows
![tableocr_pipeline](../../doc/table/tableocr_pipeline_en.jpg)
1. The coordinates of single-line text is detected by DB model, and then sends it to the recognition model to get the recognition result.
2. The table structure and cell coordinates is predicted by RARE model.
3. The recognition result of the cell is combined by the coordinates, recognition result of the single line and the coordinates of the cell.
4. The cell recognition result and the table structure together construct the html string of the table.
## 2. Performance
We evaluated the algorithm on the PubTabNet<sup>[1]</sup> eval dataset, and the performance is as follows:
|Method|[TEDS(Tree-Edit-Distance-based Similarity)](https://github.com/ibm-aur-nlp/PubTabNet/tree/master/src)|
| --- | --- |
| EDD<sup>[2]</sup> | 88.3 |
| Ours | 93.32 |
## 3. How to use
### 3.1 quick start
```python
cd PaddleOCR/ppstructure
# download model
mkdir inference && cd inference
# Download the detection model of the ultra-lightweight table English OCR model and unzip it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar && tar xf en_ppocr_mobile_v2.0_table_det_infer.tar
# Download the recognition model of the ultra-lightweight table English OCR model and unzip it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar && tar xf en_ppocr_mobile_v2.0_table_rec_infer.tar
# Download the ultra-lightweight English table inch model and unzip it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar && tar xf en_ppocr_mobile_v2.0_table_structure_infer.tar
cd ..
# run
python3 table/predict_table.py --det_model_dir=inference/en_ppocr_mobile_v2.0_table_det_infer --rec_model_dir=inference/en_ppocr_mobile_v2.0_table_rec_infer --table_model_dir=inference/en_ppocr_mobile_v2.0_table_structure_infer --image_dir=../doc/table/table.jpg --rec_char_dict_path=../ppocr/utils/ppocr_keys_v1.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=ch --det_limit_side_len=736 --det_limit_type=min --output ../output/table
```
Note: The above model is trained on the PubLayNet dataset and only supports English scanning scenarios. If you need to identify other scenarios, you need to train the model yourself and replace the three fields `det_model_dir`, `rec_model_dir`, `table_model_dir`.
After running, the excel sheet of each picture will be saved in the directory specified by the output field
### 3.2 Train
In this chapter, we only introduce the training of the table structure model, For model training of [text detection](../../doc/doc_en/detection_en.md) and [text recognition](../../doc/doc_en/recognition_en.md), please refer to the corresponding documents
#### data preparation
The training data uses public data set [PubTabNet](https://arxiv.org/abs/1911.10683 ), Can be downloaded from the official [website](https://github.com/ibm-aur-nlp/PubTabNet) 。The PubTabNet data set contains about 500,000 images, as well as annotations in html format。
#### Start training
*If you are installing the cpu version of paddle, please modify the `use_gpu` field in the configuration file to false*
```shell
# single GPU training
python3 tools/train.py -c configs/table/table_mv3.yml
# 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/table/table_mv3.yml
```
In the above instruction, use `-c` to select the training to use the `configs/table/table_mv3.yml` configuration file.
For a detailed explanation of the configuration file, please refer to [config](../../doc/doc_en/config_en.md).
#### 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.
```shell
python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./your/trained/model
```
**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.
### 3.3 Eval
The table uses [TEDS(Tree-Edit-Distance-based Similarity)](https://github.com/ibm-aur-nlp/PubTabNet/tree/master/src) as the evaluation metric of the model. Before the model evaluation, the three models in the pipeline need to be exported as inference models (we have provided them), and the gt for evaluation needs to be prepared. Examples of gt are as follows:
```json
{"PMC4289340_004_00.png": [
["<html>", "<body>", "<table>", "<thead>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</thead>", "<tbody>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</tbody>", "</table>", "</body>", "</html>"],
[[1, 4, 29, 13], [137, 4, 161, 13], [215, 4, 236, 13], [1, 17, 30, 27], [137, 17, 147, 27], [215, 17, 225, 27]],
[["<b>", "F", "e", "a", "t", "u", "r", "e", "</b>"], ["<b>", "G", "b", "3", " ", "+", "</b>"], ["<b>", "G", "b", "3", " ", "-", "</b>"], ["<b>", "P", "a", "t", "i", "e", "n", "t", "s", "</b>"], ["6", "2"], ["4", "5"]]
]}
```
In gt json, the key is the image name, the value is the corresponding gt, and gt is a list composed of four items, and each item is
1. HTML string list of table structure
2. The coordinates of each cell (not including the empty text in the cell)
3. The text information in each cell (not including the empty text in the cell)
Use the following command to evaluate. After the evaluation is completed, the teds indicator will be output.
```python
cd PaddleOCR/ppstructure
python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --gt_path=path/to/gt.json
```
If the PubLatNet eval dataset is used, it will be output
```bash
teds: 93.32
```
### 3.4 Inference
```python
cd PaddleOCR/ppstructure
python3 table/predict_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --output ../output/table
```
After running, the excel sheet of each picture will be saved in the directory specified by the output field
Reference
1. https://github.com/ibm-aur-nlp/PubTabNet
2. https://arxiv.org/pdf/1911.10683
\ No newline at end of file
test1/table/README_ch.md → ppstructure/table/README_ch.md
View file @ dad8397f
# 表格结构和内容预测 # 表格识别
## 1. pipeline ## 1. 表格识别 pipeline
表格的ocr主要包含三个模型 表格识别主要包含三个模型
1. 单行文本检测-DB 1. 单行文本检测-DB
2. 单行文本识别-CRNN 2. 单行文本识别-CRNN
3. 表格结构和cell坐标预测-RARE 3. 表格结构和cell坐标预测-RARE
具体流程图如下 具体流程图如下
![tableocr_pipeline](../../doc/table/tableocr_pipeline.png) ![tableocr_pipeline](../../doc/table/tableocr_pipeline.jpg)
1. 图片由单行文字检测检测模型到单行文字的坐标，然后送入识别模型拿到识别结果。 流程说明:
1. 图片由单行文字检测模型检测到单行文字的坐标，然后送入识别模型拿到识别结果。
2. 图片由表格结构和cell坐标预测模型拿到表格的结构信息和单元格的坐标信息。 2. 图片由表格结构和cell坐标预测模型拿到表格的结构信息和单元格的坐标信息。
3. 由单行文字的坐标、识别结果和单元格的坐标一起组合出单元格的识别结果。 3. 由单行文字的坐标、识别结果和单元格的坐标一起组合出单元格的识别结果。
4. 单元格的识别结果和表格结构一起构造表格的html字符串。 4. 单元格的识别结果和表格结构一起构造表格的html字符串。
## 2. 使用 ## 2. 性能
我们在 PubTabNet<sup>[1]</sup> 评估数据集上对算法进行了评估，性能如下
|算法|[TEDS(Tree-Edit-Distance-based Similarity)](https://github.com/ibm-aur-nlp/PubTabNet/tree/master/src)|
| --- | --- |
| EDD<sup>[2]</sup> | 88.3 |
| Ours | 93.32 |
## 3. 使用
### 3.1 快速开始
```python
cd PaddleOCR/ppstructure
# 下载模型
mkdir inference && cd inference
# 下载超轻量级表格英文OCR模型的检测模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar && tar xf en_ppocr_mobile_v2.0_table_det_infer.tar
# 下载超轻量级表格英文OCR模型的识别模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar && tar xf en_ppocr_mobile_v2.0_table_rec_infer.tar
# 下载超轻量级英文表格英寸模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar && tar xf en_ppocr_mobile_v2.0_table_structure_infer.tar
cd ..
# 执行预测
python3 table/predict_table.py --det_model_dir=inference/en_ppocr_mobile_v2.0_table_det_infer --rec_model_dir=inference/en_ppocr_mobile_v2.0_table_rec_infer --table_model_dir=inference/en_ppocr_mobile_v2.0_table_structure_infer --image_dir=../doc/table/table.jpg --rec_char_dict_path=../ppocr/utils/ppocr_keys_v1.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=ch --det_limit_side_len=736 --det_limit_type=min --output ../output/table
```
运行完成后，每张图片的excel表格会保存到output字段指定的目录下
note: 上述模型是在 PubLayNet 数据集上训练的表格识别模型，仅支持英文扫描场景，如需识别其他场景需要自己训练模型后替换 `det_model_dir`,`rec_model_dir`,`table_model_dir`三个字段即可。
### 3.2 训练
在这一章节中，我们仅介绍表格结构模型的训练，[文字检测](../../doc/doc_ch/detection.md)和[文字识别](../../doc/doc_ch/recognition.md)的模型训练请参考对应的文档。
### 2.1 训练
#### 数据准备 #### 数据准备
训练数据使用公开数据集[PubTabNet](https://arxiv.org/abs/1911.10683)，可以从[官网](https://github.com/ibm-aur-nlp/PubTabNet)下载。PubTabNet数据集包含约50万张表格数据的图像，以及图像对应的html格式的注释。 训练数据使用公开数据集PubTabNet ([论文](https://arxiv.org/abs/1911.10683)，[下载地址](https://github.com/ibm-aur-nlp/PubTabNet))。PubTabNet数据集包含约50万张表格数据的图像，以及图像对应的html格式的注释。
#### 启动训练 #### 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* *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*
...@@ -31,7 +64,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml ...@@ -31,7 +64,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/table/table_mv3.yml python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/table/table_mv3.yml
``` ```
上述指令中，通过-c 选择训练使用configs/table/table_mv3.yml配置文件。有关配置文件的详细解释，请参考[链接](./config.md)。 上述指令中，通过-c 选择训练使用configs/table/table_mv3.yml配置文件。有关配置文件的详细解释，请参考[链接](../../doc/doc_ch/config.md)。
#### 断点训练 #### 断点训练
...@@ -43,29 +76,38 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo ...@@ -43,29 +76,38 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo
**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。 **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。
### 2.2 评估 ### 3.3 评估
先cd到PaddleOCR/ppstructure目录下
表格使用 TEDS(Tree-Edit-Distance-based Similarity) 作为模型的评估指标。在进行模型评估之前，需要将pipeline中的三个模型分别导出为inference模型(我们已经提供好)，还需要准备评估的gt， gt示例如下: 表格使用 [TEDS(Tree-Edit-Distance-based Similarity)](https://github.com/ibm-aur-nlp/PubTabNet/tree/master/src) 作为模型的评估指标。在进行模型评估之前，需要将pipeline中的三个模型分别导出为inference模型(我们已经提供好)，还需要准备评估的gt， gt示例如下:
```json ```json
{"PMC4289340_004_00.png": [["<html>", "<body>", "<table>", "<thead>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</thead>", "<tbody>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</tbody>", "</table>", "</body>", "</html>"], [[1, 4, 29, 13], [137, 4, 161, 13], [215, 4, 236, 13], [1, 17, 30, 27], [137, 17, 147, 27], [215, 17, 225, 27]], [["<b>", "F", "e", "a", "t", "u", "r", "e", "</b>"], ["<b>", "G", "b", "3", " ", "+", "</b>"], ["<b>", "G", "b", "3", " ", "-", "</b>"], ["<b>", "P", "a", "t", "i", "e", "n", "t", "s", "</b>"], ["6", "2"], ["4", "5"]]]} {"PMC4289340_004_00.png": [
["<html>", "<body>", "<table>", "<thead>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</thead>", "<tbody>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</tbody>", "</table>", "</body>", "</html>"],
[[1, 4, 29, 13], [137, 4, 161, 13], [215, 4, 236, 13], [1, 17, 30, 27], [137, 17, 147, 27], [215, 17, 225, 27]],
[["<b>", "F", "e", "a", "t", "u", "r", "e", "</b>"], ["<b>", "G", "b", "3", " ", "+", "</b>"], ["<b>", "G", "b", "3", " ", "-", "</b>"], ["<b>", "P", "a", "t", "i", "e", "n", "t", "s", "</b>"], ["6", "2"], ["4", "5"]]
]}
``` ```
json 中，key为图片名，value为对于的gt，gt是一个由四个item组成的list，每个item分别为 json 中，key为图片名，value为对应的gt，gt是一个由三个item组成的list，每个item分别为
1. 表格结构的html字符串list 1. 表格结构的html字符串list
2. 每个cell的坐标 (不包括cell里文字为空的) 2. 每个cell的坐标 (不包括cell里文字为空的)
3. 每个cell里的文字信息 (不包括cell里文字为空的) 3. 每个cell里的文字信息 (不包括cell里文字为空的)
4. 每个cell里的文字信息 (包括cell里文字为空的)
准备完成后使用如下命令进行评估，评估完成后会输出teds指标。 准备完成后使用如下命令进行评估，评估完成后会输出teds指标。
```python ```python
cd PaddleOCR/ppstructure
python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --gt_path=path/to/gt.json python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --gt_path=path/to/gt.json
``` ```
如使用PubLatNet评估数据集，将会输出
```bash
teds: 93.32
```
### 3.4 预测
### 2.3 预测
先cd到PaddleOCR/ppstructure目录下
```python ```python
cd PaddleOCR/ppstructure
python3 table/predict_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --output ../output/table python3 table/predict_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --output ../output/table
``` ```
运行完成后，每张图片的excel表格会保存到output字段指定的目录下
Reference
1. https://github.com/ibm-aur-nlp/PubTabNet
2. https://arxiv.org/pdf/1911.10683
\ No newline at end of file
test1/__init__.py → ppstructure/table/__init__.py
View file @ dad8397f
...@@ -11,7 +11,3 @@ ...@@ -11,7 +11,3 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and # See the License for the specific language governing permissions and
# limitations under the License. # limitations under the License.
from .paddlestructure import PaddleStructure, draw_result, to_excel
__all__ = ['PaddleStructure', 'draw_result', 'to_excel']
test1/table/eval_table.py → ppstructure/table/eval_table.py
View file @ dad8397f
...@@ -20,9 +20,9 @@ sys.path.append(os.path.abspath(os.path.join(__dir__, '../..'))) ...@@ -20,9 +20,9 @@ sys.path.append(os.path.abspath(os.path.join(__dir__, '../..')))
import cv2 import cv2
import json import json
from tqdm import tqdm from tqdm import tqdm
from test1.table.table_metric import TEDS from ppstructure.table.table_metric import TEDS
from test1.table.predict_table import TableSystem from ppstructure.table.predict_table import TableSystem
from test1.utility import init_args from ppstructure.utility import init_args
from ppocr.utils.logging import get_logger from ppocr.utils.logging import get_logger
logger = get_logger() logger = get_logger()
...@@ -46,20 +46,20 @@ def main(gt_path, img_root, args): ...@@ -46,20 +46,20 @@ def main(gt_path, img_root, args):
pred_html = text_sys(img) pred_html = text_sys(img)
pred_htmls.append(pred_html) pred_htmls.append(pred_html)
gt_structures, gt_bboxes, gt_contents, contents_with_block = jsons_gt[img_name] gt_structures, gt_bboxes, gt_contents = jsons_gt[img_name]
gt_html, gt = get_gt_html(gt_structures, contents_with_block) gt_html, gt = get_gt_html(gt_structures, gt_contents)
gt_htmls.append(gt_html) gt_htmls.append(gt_html)
scores = teds.batch_evaluate_html(gt_htmls, pred_htmls) scores = teds.batch_evaluate_html(gt_htmls, pred_htmls)
logger.info('teds:', sum(scores) / len(scores)) logger.info('teds:', sum(scores) / len(scores))
def get_gt_html(gt_structures, contents_with_block): def get_gt_html(gt_structures, gt_contents):
end_html = [] end_html = []
td_index = 0 td_index = 0
for tag in gt_structures: for tag in gt_structures:
if '</td>' in tag: if '</td>' in tag:
if contents_with_block[td_index] != []: if gt_contents[td_index] != []:
end_html.extend(contents_with_block[td_index]) end_html.extend(gt_contents[td_index])
end_html.append(tag) end_html.append(tag)
td_index += 1 td_index += 1
else: else:
......
test1/table/predict_structure.py → ppstructure/table/predict_structure.py
View file @ dad8397f
...@@ -22,17 +22,14 @@ os.environ["FLAGS_allocator_strategy"] = 'auto_growth' ...@@ -22,17 +22,14 @@ os.environ["FLAGS_allocator_strategy"] = 'auto_growth'
import cv2 import cv2
import numpy as np import numpy as np
import math
import time import time
import traceback
import paddle
import tools.infer.utility as utility import tools.infer.utility as utility
from ppocr.data import create_operators, transform from ppocr.data import create_operators, transform
from ppocr.postprocess import build_post_process from ppocr.postprocess import build_post_process
from ppocr.utils.logging import get_logger from ppocr.utils.logging import get_logger
from ppocr.utils.utility import get_image_file_list, check_and_read_gif from ppocr.utils.utility import get_image_file_list, check_and_read_gif
from test1.utility import parse_args from ppstructure.utility import parse_args
logger = get_logger() logger = get_logger()
......
test1/table/predict_table.py → ppstructure/table/predict_table.py
View file @ dad8397f
...@@ -30,9 +30,9 @@ import tools.infer.predict_rec as predict_rec ...@@ -30,9 +30,9 @@ import tools.infer.predict_rec as predict_rec
import tools.infer.predict_det as predict_det import tools.infer.predict_det as predict_det
from ppocr.utils.utility import get_image_file_list, check_and_read_gif from ppocr.utils.utility import get_image_file_list, check_and_read_gif
from ppocr.utils.logging import get_logger from ppocr.utils.logging import get_logger
from test1.table.matcher import distance, compute_iou from ppstructure.table.matcher import distance, compute_iou
from test1.utility import parse_args from ppstructure.utility import parse_args
import test1.table.predict_structure as predict_strture import ppstructure.table.predict_structure as predict_strture
logger = get_logger() logger = get_logger()
......
test1/utility.py → ppstructure/utility.py
View file @ dad8397f
...@@ -27,7 +27,7 @@ def init_args(): ...@@ -27,7 +27,7 @@ def init_args():
parser.add_argument("--table_model_dir", type=str) parser.add_argument("--table_model_dir", type=str)
parser.add_argument("--table_char_type", type=str, default='en') parser.add_argument("--table_char_type", type=str, default='en')
parser.add_argument("--table_char_dict_path", type=str, default="../ppocr/utils/dict/table_structure_dict.txt") parser.add_argument("--table_char_dict_path", type=str, default="../ppocr/utils/dict/table_structure_dict.txt")
parser.add_argument("--layout_path_model", type=str, default="lp://PubLayNet/ppyolov2_r50vd_dcn_365e_publaynet/config")
return parser return parser
...@@ -36,15 +36,13 @@ def parse_args(): ...@@ -36,15 +36,13 @@ def parse_args():
return parser.parse_args() return parser.parse_args()
def draw_result(image, result, font_path): def draw_structure_result(image, result, font_path):
if isinstance(image, np.ndarray): if isinstance(image, np.ndarray):
image = Image.fromarray(image) image = Image.fromarray(image)
boxes, txts, scores = [], [], [] boxes, txts, scores = [], [], []
for region in result: for region in result:
if region['type'] == 'Table': if region['type'] == 'Table':
pass pass
elif region['type'] == 'Figure':
pass
else: else:
for box, rec_res in zip(region['res'][0], region['res'][1]): for box, rec_res in zip(region['res'][0], region['res'][1]):
boxes.append(np.array(box).reshape(-1, 2)) boxes.append(np.array(box).reshape(-1, 2))
......
setup.py
View file @ dad8397f
...@@ -14,6 +14,7 @@ ...@@ -14,6 +14,7 @@
from setuptools import setup from setuptools import setup
from io import open from io import open
from paddleocr import VERSION
with open('requirements.txt', encoding="utf-8-sig") as f: with open('requirements.txt', encoding="utf-8-sig") as f:
requirements = f.readlines() requirements = f.readlines()
...@@ -32,7 +33,7 @@ setup( ...@@ -32,7 +33,7 @@ setup(
package_dir={'paddleocr': ''}, package_dir={'paddleocr': ''},
include_package_data=True, include_package_data=True,
entry_points={"console_scripts": ["paddleocr= paddleocr.paddleocr:main"]}, entry_points={"console_scripts": ["paddleocr= paddleocr.paddleocr:main"]},
version='2.0.6', version=VERSION,
install_requires=requirements, install_requires=requirements,
license='Apache License 2.0', license='Apache License 2.0',
description='Awesome OCR toolkits based on PaddlePaddle （8.6M ultra-lightweight pre-trained model, support training and deployment among server, mobile, embeded and IoT devices', description='Awesome OCR toolkits based on PaddlePaddle （8.6M ultra-lightweight pre-trained model, support training and deployment among server, mobile, embeded and IoT devices',
......
test1/MANIFEST.in deleted 100644 → 0
View file @ 5801311c
include LICENSE
include README.md
recursive-include ppocr/utils *.txt utility.py logging.py network.py
recursive-include ppocr/data/ *.py
recursive-include ppocr/postprocess *.py
recursive-include tools/infer *.py
recursive-include test1 *.py
test1/api.md deleted 100644 → 0
View file @ 5801311c
# PaddleStructure
install layoutparser
```sh
wget https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
pip3 install layoutparser-0.0.0-py3-none-any.whl
```
## 1. Introduction to pipeline
PaddleStructure is a toolkit for complex layout text OCR, the process is as follows
![pipeline](../doc/table/pipeline.png)
In PaddleStructure, the image will be analyzed by layoutparser first. In the layout analysis, the area in the image will be classified, and the OCR process will be carried out according to the category.
Currently layoutparser will output five categories:
1. Text
2. Title
3. Figure
4. List
5. Table
Types 1-4 follow the traditional OCR process, and 5 follow the Table OCR process.
## 2. LayoutParser
## 3. Table OCR
[doc](table/README.md)
## 4. Predictive by inference engine
Use the following commands to complete the inference
```python
python3 table/predict_system.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --output ../output/table
```
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.
## 5. PaddleStructure whl package introduction
### 5.1 Use
5.1.1 Use by code
```python
import os
import cv2
from paddlestructure import PaddleStructure,draw_result,save_res
table_engine = PaddleStructure(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])
for line in result:
print(line)
from PIL import Image
font_path = 'path/tp/PaddleOCR/doc/fonts/simfang.ttf'
image = Image.open(img_path).convert('RGB')
im_show = draw_result(image, result,font_path=font_path)
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
5.1.2 Use by command line
```bash
paddlestructure --image_dir=../doc/table/1.png
```
### Parameter Description
Most of the parameters are consistent with the paddleocr whl package, see [whl package documentation](../doc/doc_ch/whl.md)
| Parameter | Description | Default |
|------------------------|------------------------------------------------------|------------------|
| output | The path where excel and recognition results are saved | ./output/table |
| structure_max_len | When the table structure model predicts, the long side of the image is resized | 488 |
| structure_model_dir | Table structure inference model path | None |
| structure_char_type | Dictionary path used by table structure model | ../ppocr/utils/dict/table_structure_dict.tx |
test1/api_ch.md deleted 100644 → 0
View file @ 5801311c
# PaddleStructure
安装layoutparser
```sh
wget https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
pip3 install layoutparser-0.0.0-py3-none-any.whl
```
## 1. pipeline介绍
PaddleStructure 是一个用于复杂板式文字OCR的工具包，流程如下
![pipeline](../doc/table/pipeline.png)
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，根据根据类别进行对于的ocr流程。
目前layoutparser会输出五个类别:
1. Text
2. Title
3. Figure
4. List
5. Table
1-4类走传统的OCR流程，5走表格的OCR流程。
## 2. LayoutParser
[文档](layout/README.md)
## 3. Table OCR
[文档](table/README_ch.md)
## 4. 预测引擎推理
使用如下命令即可完成预测引擎的推理
```python
python3 table/predict_system.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --rec_char_type=EN --det_limit_side_len=736 --det_limit_type=min --output ../output/table
```
运行完成后，每张图片会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。
## 5. PaddleStructure whl包介绍
### 5.1 使用
5.1.1 代码使用
```python
import os
import cv2
from paddlestructure import PaddleStructure,draw_result,save_res
table_engine = PaddleStructure(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])
for line in result:
print(line)
from PIL import Image
font_path = 'path/tp/PaddleOCR/doc/fonts/simfang.ttf'
image = Image.open(img_path).convert('RGB')
im_show = draw_result(image, result,font_path=font_path)
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
5.1.2 命令行使用
```bash
paddlestructure --image_dir=../doc/table/1.png
```
### 参数说明
大部分参数和paddleocr whl包保持一致，见 [whl包文档](../doc/doc_ch/whl.md)
| 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.tx |
test1/paddlestructure.py deleted 100644 → 0
View file @ 5801311c
# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import os
import sys
__dir__ = os.path.dirname(__file__)
sys.path.append(__dir__)
sys.path.append(os.path.join(__dir__, '..'))
import cv2
import numpy as np
from pathlib import Path
from ppocr.utils.logging import get_logger
from test1.predict_system import OCRSystem, save_res
from test1.table.predict_table import to_excel
from test1.utility import init_args, draw_result
logger = get_logger()
from ppocr.utils.utility import check_and_read_gif, get_image_file_list
from ppocr.utils.network import maybe_download, download_with_progressbar, confirm_model_dir_url, is_link
__all__ = ['PaddleStructure', 'draw_result', 'save_res']
VERSION = '2.1'
BASE_DIR = os.path.expanduser("~/.paddlestructure/")
model_urls = {
'det': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar',
'rec': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar',
'table': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar'
}
def parse_args(mMain=True):
import argparse
parser = init_args()
parser.add_help = mMain
for action in parser._actions:
if action.dest in ['rec_char_dict_path', 'table_char_dict_path']:
action.default = None
if mMain:
return parser.parse_args()
else:
inference_args_dict = {}
for action in parser._actions:
inference_args_dict[action.dest] = action.default
return argparse.Namespace(**inference_args_dict)
class PaddleStructure(OCRSystem):
def __init__(self, **kwargs):
params = parse_args(mMain=False)
params.__dict__.update(**kwargs)
if not params.show_log:
logger.setLevel(logging.INFO)
params.use_angle_cls = False
# init model dir
params.det_model_dir, det_url = confirm_model_dir_url(params.det_model_dir,
os.path.join(BASE_DIR, VERSION, 'det'),
model_urls['det'])
params.rec_model_dir, rec_url = confirm_model_dir_url(params.rec_model_dir,
os.path.join(BASE_DIR, VERSION, 'rec'),
model_urls['rec'])
params.table_model_dir, table_url = confirm_model_dir_url(params.table_model_dir,
os.path.join(BASE_DIR, VERSION, 'table'),
model_urls['table'])
# 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_type = 'EN'
if os.path.exists(str(Path(__file__).parent / 'ppocr/utils/dict/table_dict.txt')):
params.rec_char_dict_path = str(Path(__file__).parent / 'ppocr/utils/dict/table_dict.txt')
else:
params.rec_char_dict_path = str(Path(__file__).parent.parent / 'ppocr/utils/dict/table_dict.txt')
if params.table_char_dict_path is None:
if os.path.exists(str(Path(__file__).parent / 'ppocr/utils/dict/table_structure_dict.txt')):
params.table_char_dict_path = str(
Path(__file__).parent / 'ppocr/utils/dict/table_structure_dict.txt')
else:
params.table_char_dict_path = str(
Path(__file__).parent.parent / 'ppocr/utils/dict/table_structure_dict.txt')
print(params)
super().__init__(params)
def __call__(self, img):
if isinstance(img, str):
# download net image
if img.startswith('http'):
download_with_progressbar(img, 'tmp.jpg')
img = 'tmp.jpg'
image_file = img
img, flag = check_and_read_gif(image_file)
if not flag:
with open(image_file, 'rb') as f:
np_arr = np.frombuffer(f.read(), dtype=np.uint8)
img = cv2.imdecode(np_arr, cv2.IMREAD_COLOR)
if img is None:
logger.error("error in loading image:{}".format(image_file))
return None
if isinstance(img, np.ndarray) and len(img.shape) == 2:
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
res = super().__call__(img)
return res
def main():
# for cmd
args = parse_args(mMain=True)
image_dir = args.image_dir
save_folder = args.output
if image_dir.startswith('http'):
download_with_progressbar(image_dir, 'tmp.jpg')
image_file_list = ['tmp.jpg']
else:
image_file_list = get_image_file_list(args.image_dir)
if len(image_file_list) == 0:
logger.error('no images find in {}'.format(args.image_dir))
return
structure_engine = PaddleStructure(**(args.__dict__))
for img_path in image_file_list:
img_name = os.path.basename(img_path).split('.')[0]
logger.info('{}{}{}'.format('*' * 10, img_path, '*' * 10))
result = structure_engine(img_path)
for item in result:
logger.info(item['res'])
save_res(result, save_folder, img_name)
logger.info('result save to {}'.format(os.path.join(save_folder, img_name)))
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment