Skip to content

GitLab

Unverified Commit 5d35c733 authored by Daniel Yang's avatar Daniel Yang Committed by GitHub
Browse files

Merge pull request #3497 from WenmuZhou/table2 

merge paddlestructure whl to paddleocr whl
parents c86ee9fe fda89b35
Show whitespace changes
Inline Side-by-side
Showing with 333 additions and 362 deletions
MANIFEST.in
View file @ 5d35c733
...@@ -6,3 +6,4 @@ recursive-include ppocr/data *.py ...@@ -6,3 +6,4 @@ recursive-include ppocr/data *.py
recursive-include ppocr/postprocess *.py recursive-include ppocr/postprocess *.py
recursive-include tools/infer *.py recursive-include tools/infer *.py
recursive-include ppocr/utils/e2e_utils *.py recursive-include ppocr/utils/e2e_utils *.py
recursive-include ppstructure *.py
\ No newline at end of file
__init__.py
View file @ 5d35c733
...@@ -11,7 +11,8 @@ ...@@ -11,7 +11,8 @@
# 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.
import paddleocr
from .paddleocr import *
__all__ = ['PaddleOCR', 'draw_ocr'] __version__ = paddleocr.VERSION
from .paddleocr import PaddleOCR __all__ = ['PaddleOCR', 'PPStructure', 'draw_ocr', 'draw_structure_result', 'save_structure_res']
from .tools.infer.utility import draw_ocr
doc/doc_ch/whl.md
View file @ 5d35c733
...@@ -356,3 +356,4 @@ im_show.save('result.jpg') ...@@ -356,3 +356,4 @@ im_show.save('result.jpg')
| rec | 前向时是否启动识别 | TRUE | | 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 | | 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 | | 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 |
doc/doc_en/whl_en.md
View file @ 5d35c733
...@@ -362,5 +362,5 @@ im_show.save('result.jpg') ...@@ -362,5 +362,5 @@ im_show.save('result.jpg')
| det | Enable detction when `ppocr.ocr` func exec | TRUE | | det | Enable detction when `ppocr.ocr` func exec | TRUE |
| rec | Enable recognition 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 | | 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 | show_log | Whether to print log in det and rec | FALSE |
| FALSE | | type | Perform ocr or table structuring, the value is selected in ['ocr','structure'] | ocr |
\ No newline at end of file \ No newline at end of file
paddleocr.py
View file @ 5d35c733
...@@ -29,16 +29,19 @@ from ppocr.utils.logging import get_logger ...@@ -29,16 +29,19 @@ from ppocr.utils.logging import get_logger
logger = get_logger() logger = get_logger()
from ppocr.utils.utility import check_and_read_gif, get_image_file_list from ppocr.utils.utility import check_and_read_gif, get_image_file_list
from ppocr.utils.network import maybe_download, download_with_progressbar, is_link, confirm_model_dir_url from ppocr.utils.network import maybe_download, download_with_progressbar, is_link, confirm_model_dir_url
from tools.infer.utility import draw_ocr, init_args, str2bool 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'] __all__ = ['PaddleOCR', 'PPStructure', 'draw_ocr', 'draw_structure_result', 'save_structure_res']
model_urls = { model_urls = {
'det': { 'det': {
'ch': 'ch':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar', 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar',
'en': 'en':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_ppocr_mobile_v2.0_det_infer.tar' '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'
}, },
'rec': { 'rec': {
'ch': { 'ch': {
...@@ -110,14 +113,21 @@ model_urls = { ...@@ -110,14 +113,21 @@ model_urls = {
'url': 'url':
'https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/devanagari_ppocr_mobile_v2.0_rec_infer.tar', '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': 'cls': 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar',
'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'] SUPPORT_DET_MODEL = ['DB']
VERSION = '2.1' VERSION = '2.2'
SUPPORT_REC_MODEL = ['CRNN'] SUPPORT_REC_MODEL = ['CRNN']
BASE_DIR = os.path.expanduser("~/.paddleocr/") BASE_DIR = os.path.expanduser("~/.paddleocr/")
...@@ -129,9 +139,10 @@ def parse_args(mMain=True): ...@@ -129,9 +139,10 @@ def parse_args(mMain=True):
parser.add_argument("--lang", type=str, default='ch') parser.add_argument("--lang", type=str, default='ch')
parser.add_argument("--det", type=str2bool, default=True) parser.add_argument("--det", type=str2bool, default=True)
parser.add_argument("--rec", type=str2bool, default=True) parser.add_argument("--rec", type=str2bool, default=True)
parser.add_argument("--type", type=str, default='ocr')
for action in parser._actions: for action in parser._actions:
if action.dest == 'rec_char_dict_path': if action.dest in ['rec_char_dict_path', 'table_char_dict_path']:
action.default = None action.default = None
if mMain: if mMain:
return parser.parse_args() return parser.parse_args()
...@@ -142,19 +153,7 @@ def parse_args(mMain=True): ...@@ -142,19 +153,7 @@ def parse_args(mMain=True):
return argparse.Namespace(**inference_args_dict) return argparse.Namespace(**inference_args_dict)
class PaddleOCR(predict_system.TextSystem): def parse_lang(lang):
def __init__(self, **kwargs):
"""
paddleocr package
args:
**kwargs: other params show in paddleocr --help
"""
params = parse_args(mMain=False)
params.__dict__.update(**kwargs)
if not params.show_log:
logger.setLevel(logging.INFO)
self.use_angle_cls = params.use_angle_cls
lang = params.lang
latin_lang = [ latin_lang = [
'af', 'az', 'bs', 'cs', 'cy', 'da', 'de', 'es', 'et', 'fr', 'ga', 'af', 'az', 'bs', 'cs', 'cy', 'da', 'de', 'es', 'et', 'fr', 'ga',
'hr', 'hu', 'id', 'is', 'it', 'ku', 'la', 'lt', 'lv', 'mi', 'ms', 'hr', 'hu', 'id', 'is', 'it', 'ku', 'la', 'lt', 'lv', 'mi', 'ms',
...@@ -183,23 +182,36 @@ class PaddleOCR(predict_system.TextSystem): ...@@ -183,23 +182,36 @@ class PaddleOCR(predict_system.TextSystem):
model_urls['rec'].keys(), lang) model_urls['rec'].keys(), lang)
if lang == "ch": if lang == "ch":
det_lang = "ch" det_lang = "ch"
elif lang == 'structure':
det_lang = 'structure'
else: else:
det_lang = "en" det_lang = "en"
use_inner_dict = False return lang, det_lang
if params.rec_char_dict_path is None:
use_inner_dict = True
params.rec_char_dict_path = model_urls['rec'][lang][ class PaddleOCR(predict_system.TextSystem):
'dict_path'] def __init__(self, **kwargs):
"""
paddleocr package
args:
**kwargs: other params show in paddleocr --help
"""
params = parse_args(mMain=False)
params.__dict__.update(**kwargs)
if not params.show_log:
logger.setLevel(logging.INFO)
self.use_angle_cls = params.use_angle_cls
lang, det_lang = parse_lang(params.lang)
# init model dir # init model dir
params.det_model_dir, det_url = confirm_model_dir_url(params.det_model_dir, params.det_model_dir, det_url = confirm_model_dir_url(params.det_model_dir,
os.path.join(BASE_DIR, VERSION, 'det', det_lang), os.path.join(BASE_DIR, VERSION, 'ocr', 'det', det_lang),
model_urls['det'][det_lang]) model_urls['det'][det_lang])
params.rec_model_dir, rec_url = confirm_model_dir_url(params.rec_model_dir, params.rec_model_dir, rec_url = confirm_model_dir_url(params.rec_model_dir,
os.path.join(BASE_DIR, VERSION, 'rec', lang), os.path.join(BASE_DIR, VERSION, 'ocr', 'rec', lang),
model_urls['rec'][lang]['url']) model_urls['rec'][lang]['url'])
params.cls_model_dir, cls_url = confirm_model_dir_url(params.cls_model_dir, params.cls_model_dir, cls_url = confirm_model_dir_url(params.cls_model_dir,
os.path.join(BASE_DIR, VERSION, 'cls'), os.path.join(BASE_DIR, VERSION, 'ocr', 'cls'),
model_urls['cls']) model_urls['cls'])
# download model # download model
maybe_download(params.det_model_dir, det_url) maybe_download(params.det_model_dir, det_url)
...@@ -212,9 +224,9 @@ class PaddleOCR(predict_system.TextSystem): ...@@ -212,9 +224,9 @@ class PaddleOCR(predict_system.TextSystem):
if params.rec_algorithm not in SUPPORT_REC_MODEL: if params.rec_algorithm not in SUPPORT_REC_MODEL:
logger.error('rec_algorithm must in {}'.format(SUPPORT_REC_MODEL)) logger.error('rec_algorithm must in {}'.format(SUPPORT_REC_MODEL))
sys.exit(0) sys.exit(0)
if use_inner_dict:
params.rec_char_dict_path = str( if params.rec_char_dict_path is None:
Path(__file__).parent / params.rec_char_dict_path) params.rec_char_dict_path = str(Path(__file__).parent / model_urls['rec'][lang]['dict_path'])
print(params) print(params)
# init det_model and rec_model # init det_model and rec_model
...@@ -272,6 +284,59 @@ class PaddleOCR(predict_system.TextSystem): ...@@ -272,6 +284,59 @@ class PaddleOCR(predict_system.TextSystem):
return rec_res return rec_res
class PPStructure(OCRSystem):
def __init__(self, **kwargs):
params = parse_args(mMain=False)
params.__dict__.update(**kwargs)
if not params.show_log:
logger.setLevel(logging.INFO)
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'])
# 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'])
if params.table_char_dict_path is None:
params.table_char_dict_path = str(Path(__file__).parent / model_urls['table']['dict_path'])
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(): def main():
# for cmd # for cmd
args = parse_args(mMain=True) args = parse_args(mMain=True)
...@@ -284,14 +349,28 @@ def main(): ...@@ -284,14 +349,28 @@ def main():
if len(image_file_list) == 0: if len(image_file_list) == 0:
logger.error('no images find in {}'.format(args.image_dir)) logger.error('no images find in {}'.format(args.image_dir))
return return
if args.type == 'ocr':
engine = PaddleOCR(**(args.__dict__))
elif args.type == 'structure':
engine = PPStructure(**(args.__dict__))
else:
raise NotImplementedError
ocr_engine = PaddleOCR(**(args.__dict__))
for img_path in image_file_list: for img_path in image_file_list:
img_name = os.path.basename(img_path).split('.')[0]
logger.info('{}{}{}'.format('*' * 10, img_path, '*' * 10)) logger.info('{}{}{}'.format('*' * 10, img_path, '*' * 10))
result = ocr_engine.ocr(img_path, if args.type == 'ocr':
result = engine.ocr(img_path,
det=args.det, det=args.det,
rec=args.rec, rec=args.rec,
cls=args.use_angle_cls) cls=args.use_angle_cls)
if result is not None: if result is not None:
for line in result: for line in result:
logger.info(line) logger.info(line)
elif args.type == 'structure':
result = engine(img_path)
save_structure_res(result, args.output, img_name)
for item in result:
item.pop('img')
logger.info(item['res'])
\ No newline at end of file
ppocr/data/imaug/label_ops.py
View file @ 5d35c733
...@@ -380,14 +380,14 @@ class TableLabelEncode(object): ...@@ -380,14 +380,14 @@ class TableLabelEncode(object):
list_elem = [] list_elem = []
with open(character_dict_path, "rb") as fin: with open(character_dict_path, "rb") as fin:
lines = fin.readlines() lines = fin.readlines()
substr = lines[0].decode('utf-8').strip("\n").split("\t") substr = lines[0].decode('utf-8').strip("\r\n").split("\t")
character_num = int(substr[0]) character_num = int(substr[0])
elem_num = int(substr[1]) elem_num = int(substr[1])
for cno in range(1, 1+character_num): for cno in range(1, 1+character_num):
character = lines[cno].decode('utf-8').strip("\n") character = lines[cno].decode('utf-8').strip("\r\n")
list_character.append(character) list_character.append(character)
for eno in range(1+character_num, 1+character_num+elem_num): for eno in range(1+character_num, 1+character_num+elem_num):
elem = lines[eno].decode('utf-8').strip("\n") elem = lines[eno].decode('utf-8').strip("\r\n")
list_elem.append(elem) list_elem.append(elem)
return list_character, list_elem return list_character, list_elem
......
ppstructure/MANIFEST.in deleted 100644 → 0
View file @ c86ee9fe
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 ppstructure *.py
ppstructure/README.md
View file @ 5d35c733
# 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 ## 1. Quick start
### install ### install
**install PaddlePaddle2.0**
**install layoutparser** ```bash
```sh pip3 install --upgrade pip
pip3 install -U premailer paddleocr https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
# If you have cuda9 or cuda10 installed on your machine, please run the following command to install
python3 -m pip install paddlepaddle-gpu==2.0.0 -i https://mirror.baidu.com/pypi/simple
# If you only have cpu on your machine, please run the following command to install
python3 -m pip install paddlepaddle==2.0.0 -i https://mirror.baidu.com/pypi/simple
For more version requirements, please refer to the instructions in the [installation document](https://www.paddlepaddle.org.cn/install/quick) .
``` ```
**install paddlestructure**
install by pypi **Clone PaddleOCR repo**
```bash ```bash
pip install paddlestructure # 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.
```
**install paddleocr**
install by pypi
```bash
cd PaddleOCR
pip install "paddleocr>=2.2" # # Recommend to use version 2.2
``` ```
build own whl package and install build own whl package and install
```bash ```bash
python3 setup.py bdist_wheel python3 setup.py bdist_wheel
pip3 install dist/paddlestructure-x.x.x-py3-none-any.whl # x.x.x is the version of paddlestructure pip3 install dist/paddleocr-x.x.x-py3-none-any.whl # x.x.x is the version of paddleocr
```
**install layoutparser**
```sh
pip3 install -U premailer https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
``` ```
### 1.2 Use ### 1.2 Use
...@@ -28,7 +55,7 @@ pip3 install dist/paddlestructure-x.x.x-py3-none-any.whl # x.x.x is the version ...@@ -28,7 +55,7 @@ pip3 install dist/paddlestructure-x.x.x-py3-none-any.whl # x.x.x is the version
#### 1.2.1 Use by command line #### 1.2.1 Use by command line
```bash ```bash
paddlestructure --image_dir=../doc/table/1.png paddleocr --image_dir=../doc/table/1.png --type=structure
``` ```
#### 1.2.2 Use by code #### 1.2.2 Use by code
...@@ -36,29 +63,30 @@ paddlestructure --image_dir=../doc/table/1.png ...@@ -36,29 +63,30 @@ paddlestructure --image_dir=../doc/table/1.png
```python ```python
import os import os
import cv2 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' save_folder = './output/table'
img_path = '../doc/table/1.png' img_path = '../doc/table/1.png'
img = cv2.imread(img_path) img = cv2.imread(img_path)
result = table_engine(img) 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: for line in result:
line.pop('img')
print(line) print(line)
from PIL import Image 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') 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 = Image.fromarray(im_show)
im_show.save('result.jpg') im_show.save('result.jpg')
``` ```
#### 1.2.3 返回结果说明 #### 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 ```shell
[ [
...@@ -89,20 +117,20 @@ The description of each field in dict is as follows ...@@ -89,20 +117,20 @@ The description of each field in dict is as follows
Most of the parameters are consistent with the paddleocr whl package, see [doc of whl](../doc/doc_en/whl_en.md) Most of the parameters are consistent with the paddleocr whl package, see [doc of whl](../doc/doc_en/whl_en.md)
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. 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.
## 2. PaddleStructure Pipeline ## 2. PPStructure Pipeline
the process is as follows the process is as follows
![pipeline](../doc/table/pipeline_en.jpg) ![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 ### 2.1 LayoutParser
Layout analysis divides the document data into regions, including the use of Python scripts for layout analysis tools, extraction of special category detection boxes, performance indicators, and custom training layout analysis models. For details, please refer to [document](layout/README_en.md). Layout analysis divides the document data into regions, including the use of Python scripts for layout analysis tools, extraction of special category detection boxes, performance indicators, and custom training layout analysis models. For details, please refer to [document](layout/README_en.md).
### 2.2 Table OCR ### 2.2 Table Structure
Table OCR converts table image into excel documents, which include the detection and recognition of table text and the prediction of table structure and cell coordinates. For detailed, please refer to [document](table/README.md) Table OCR converts table image into excel documents, which include the detection and recognition of table text and the prediction of table structure and cell coordinates. For detailed, please refer to [document](table/README.md)
...@@ -111,15 +139,25 @@ Table OCR converts table image into excel documents, which include the detection ...@@ -111,15 +139,25 @@ Table OCR converts table image into excel documents, which include the detection
Use the following commands to complete the inference. Use the following commands to complete the inference.
```python ```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 --vis_font_path=../doc/fonts/simfang.ttf cd PaddleOCR/ppstructure
# download model
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 table structure model of the ultra-lightweight Chinese OCR model and uncompress 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 ..
python3 table/predict_system.py --det_model_dir=inference/ch_ppocr_mobile_v2.0_det_infer --rec_model_dir=inference/ch_ppocr_mobile_v2.0_rec_infer --table_model_dir=inference/en_ppocr_mobile_v2.0_table_structure_infer --image_dir=../doc/table/1.png --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 --vis_font_path=../doc/fonts/simfang.ttf
``` ```
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. 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.
**Model List** **Model List**
|model name|description|config|model size|download| |model name|description|config|model size|download|
| --- | --- | --- | --- | --- | | --- | --- | --- | --- | --- |
|en_ppocr_mobile_v2.0_table_det|Text detection in English table scene|[ch_det_mv3_db_v2.0.yml](../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)| 4.7M |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar) |
|en_ppocr_mobile_v2.0_table_rec|Text recognition in English table scene|[rec_chinese_lite_train_v2.0.yml](..//configs/rec/rec_mv3_none_bilstm_ctc.yml)|6.9M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar) |
|en_ppocr_mobile_v2.0_table_structure|Table structure prediction for English table scenarios|[table_mv3.yml](../configs/table/table_mv3.yml)|18.6M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar) | |en_ppocr_mobile_v2.0_table_structure|Table structure prediction for English table scenarios|[table_mv3.yml](../configs/table/table_mv3.yml)|18.6M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar) |
\ No newline at end of file
ppstructure/README_ch.md
View file @ 5d35c733
# 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 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 +6,59 @@ PaddleStructure是一个用于复杂版面分析的OCR工具包，其能够对 ...@@ -6,29 +6,59 @@ 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 ### 1.1 安装
**安装 layoutparser** **安装PaddlePaddle2.0**
```sh
pip3 install -U premailer paddleocr https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl ```bash
pip3 install --upgrade pip
# 如果您的机器安装的是CUDA9或CUDA10，请运行以下命令安装
python3 -m pip install paddlepaddle-gpu==2.0.0 -i https://mirror.baidu.com/pypi/simple
# 如果您的机器是CPU，请运行以下命令安装
python3 -m pip install paddlepaddle==2.0.0 -i https://mirror.baidu.com/pypi/simple
# 更多的版本需求，请参照[安装文档](https://www.paddlepaddle.org.cn/install/quick)中的说明进行操作。
``` ```
**安装 paddlestructure**
**克隆PaddleOCR repo代码**
```bash
【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。
```
**安装 paddleocr**
pip安装 pip安装
```bash ```bash
pip install paddlestructure cd PaddleOCR
pip install "paddleocr>=2.0.1" # 推荐使用2.0.1+版本
``` ```
本地构建并安装 本地构建并安装
```bash ```bash
python3 setup.py bdist_wheel 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 pip3 install dist/paddleocr-x.x.x-py3-none-any.whl # x.x.x是paddleocr的版本号
```
**安装 layoutparser**
```sh
pip3 install -U premailer paddleocr https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
``` ```
### 1.2 PaddleStructure whl包使用 ### 1.2 PPStructure whl包使用
#### 1.2.1 命令行使用 #### 1.2.1 命令行使用
```bash ```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 #### 1.2.2 Python脚本使用
...@@ -36,24 +66,25 @@ paddlestructure --image_dir=../doc/table/1.png ...@@ -36,24 +66,25 @@ paddlestructure --image_dir=../doc/table/1.png
```python ```python
import os import os
import cv2 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' save_folder = './output/table'
img_path = '../doc/table/1.png' img_path = '../doc/table/1.png'
img = cv2.imread(img_path) img = cv2.imread(img_path)
result = table_engine(img) 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: for line in result:
line.pop('img')
print(line) print(line)
from PIL import Image from PIL import Image
font_path = '../doc/fonts/simfang.ttf' # PaddleOCR下提供字体包 font_path = '../doc/fonts/simfang.ttf' # PaddleOCR下提供字体包
image = Image.open(img_path).convert('RGB') 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 = Image.fromarray(im_show)
im_show.save('result.jpg') im_show.save('result.jpg')
``` ```
...@@ -90,21 +121,21 @@ dict 里各个字段说明如下 ...@@ -90,21 +121,21 @@ dict 里各个字段说明如下
大部分参数和paddleocr whl包保持一致，见 [whl包文档](../doc/doc_ch/whl.md) 大部分参数和paddleocr whl包保持一致，见 [whl包文档](../doc/doc_ch/whl.md)
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。 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文件和图片名名为表格在图片里的坐标。
## 2. PaddleStructure Pipeline ## 2. PPStructure Pipeline
流程如下 流程如下
![pipeline](../doc/table/pipeline.jpg) ![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 ### 2.1 版面分析
版面分析对文档数据进行区域分类，其中包括版面分析工具的Python脚本使用、提取指定类别检测框、性能指标以及自定义训练版面分析模型，详细内容可以参考[文档](layout/README.md)。 版面分析对文档数据进行区域分类，其中包括版面分析工具的Python脚本使用、提取指定类别检测框、性能指标以及自定义训练版面分析模型，详细内容可以参考[文档](layout/README.md)。
### 2.2 表格识别 ### 2.2 表格结构化
Table OCR将表格图片转换为excel文档，其中包含对于表格文本的检测和识别以及对于表格结构和单元格坐标的预测，详细说明参考[文档](table/README_ch.md) Table OCR将表格图片转换为excel文档，其中包含对于表格文本的检测和识别以及对于表格结构和单元格坐标的预测，详细说明参考[文档](table/README_ch.md)
...@@ -113,14 +144,24 @@ Table OCR将表格图片转换为excel文档，其中包含对于表格文本的 ...@@ -113,14 +144,24 @@ Table OCR将表格图片转换为excel文档，其中包含对于表格文本的
使用如下命令即可完成预测引擎的推理 使用如下命令即可完成预测引擎的推理
```python ```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 --vis_font_path=../doc/fonts/simfang.ttf cd ppstructure
# 下载模型
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 ..
python3 table/predict_system.py --det_model_dir=inference/ch_ppocr_mobile_v2.0_det_infer --rec_model_dir=inference/ch_ppocr_mobile_v2.0_rec_infer --table_model_dir=inference/en_ppocr_mobile_v2.0_table_structure_infer --image_dir=../doc/table/1.png --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 --vis_font_path=../doc/fonts/simfang.ttf
``` ```
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。 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文件和图片名名为表格在图片里的坐标。
**Model List** **Model List**
|模型名称|模型简介|配置文件|推理模型大小|下载地址| |模型名称|模型简介|配置文件|推理模型大小|下载地址|
| --- | --- | --- | --- | --- | | --- | --- | --- | --- | --- |
|en_ppocr_mobile_v2.0_table_det|英文表格场景的文字检测|[ch_det_mv3_db_v2.0.yml](../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)| 4.7M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar) |
|en_ppocr_mobile_v2.0_table_rec|英文表格场景的文字识别|[rec_chinese_lite_train_v2.0.yml](../configs/rec/rec_mv3_none_bilstm_ctc.yml)|6.9M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar) |
|en_ppocr_mobile_v2.0_table_structure|英文表格场景的表格结构预测|[table_mv3.yml](../configs/table/table_mv3.yml)|18.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar) | |en_ppocr_mobile_v2.0_table_structure|英文表格场景的表格结构预测|[table_mv3.yml](../configs/table/table_mv3.yml)|18.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar) |
\ No newline at end of file
ppstructure/__init__.py
View file @ 5d35c733
...@@ -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, save_res
__all__ = ['PaddleStructure', 'draw_result', 'save_res']
ppstructure/paddlestructure.py deleted 100644 → 0
View file @ c86ee9fe
# 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 ppstructure.predict_system import OCRSystem, save_res
from ppstructure.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)))
\ No newline at end of file
ppstructure/predict_system.py
View file @ 5d35c733
...@@ -26,19 +26,18 @@ import numpy as np ...@@ -26,19 +26,18 @@ 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 ppstructure.table.predict_table import TableSystem, to_excel from ppstructure.table.predict_table import TableSystem, to_excel
from ppstructure.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):
import layoutparser as lp
args.det_limit_type = 'resize_long' args.det_limit_type = 'resize_long'
args.drop_score = 0 args.drop_score = 0
if not args.show_log: if not args.show_log:
...@@ -66,21 +65,21 @@ class OCRSystem(object): ...@@ -66,21 +65,21 @@ class OCRSystem(object):
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 # remove style char
style_token = ['<strike>','<strike>','<sup>','</sub>','<b>','</b>','<sub>','</sup>', style_token = ['<strike>', '<strike>', '<sup>', '</sub>', '<b>', '</b>', '<sub>', '</sup>',
'<overline>','</overline>','<underline>','</underline>','<i>','</i>'] '<overline>', '</overline>', '<underline>', '</underline>', '<i>', '</i>']
filter_rec_res_tmp = [] filter_rec_res_tmp = []
for rec_res in filter_rec_res: for rec_res in filter_rec_res:
rec_str, rec_conf = rec_res rec_str, rec_conf = rec_res
for token in style_token: for token in style_token:
if token in rec_str: if token in rec_str:
rec_str = rec_str.replace(token, '') rec_str = rec_str.replace(token, '')
filter_rec_res_tmp.append((rec_str,rec_conf)) filter_rec_res_tmp.append((rec_str, rec_conf))
res = (filter_boxes, filter_rec_res_tmp) res = (filter_boxes, filter_rec_res_tmp)
res_list.append({'type': region.type, 'bbox': [x1, y1, x2, y2], 'res': res}) 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
...@@ -89,6 +88,10 @@ def save_res(res, save_folder, img_name): ...@@ -89,6 +88,10 @@ def save_res(res, save_folder, img_name):
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)
if region['type'] == 'Figure':
roi_img = region['img']
img_path = os.path.join(excel_save_folder, '{}.jpg'.format(region['bbox']))
cv2.imwrite(img_path, roi_img)
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]):
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))
...@@ -115,8 +118,8 @@ def main(args): ...@@ -115,8 +118,8 @@ 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/setup.py deleted 100644 → 0
View file @ c86ee9fe
# 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 os
from setuptools import setup
from io import open
import shutil
with open('../requirements.txt', encoding="utf-8-sig") as f:
requirements = f.readlines()
requirements.append('tqdm')
def readme():
with open('README_ch.md', encoding="utf-8-sig") as f:
README = f.read()
return README
shutil.copytree('./table', './ppstructure/table')
shutil.copyfile('./predict_system.py', './ppstructure/predict_system.py')
shutil.copyfile('./utility.py', './ppstructure/utility.py')
shutil.copytree('../ppocr', './ppocr')
shutil.copytree('../tools', './tools')
shutil.copyfile('../LICENSE', './LICENSE')
setup(
name='paddlestructure',
packages=['paddlestructure'],
package_dir={'paddlestructure': ''},
include_package_data=True,
entry_points={"console_scripts": ["paddlestructure= paddlestructure.paddlestructure:main"]},
version='1.0',
install_requires=requirements,
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',
long_description=readme(),
long_description_content_type='text/markdown',
url='https://github.com/PaddlePaddle/PaddleOCR',
download_url='https://github.com/PaddlePaddle/PaddleOCR.git',
keywords=[
'ocr textdetection textrecognition paddleocr crnn east star-net rosetta ocrlite db chineseocr chinesetextdetection chinesetextrecognition'
],
classifiers=[
'Intended Audience :: Developers', 'Operating System :: OS Independent',
'Natural Language :: Chinese (Simplified)',
'Programming Language :: Python :: 3',
'Programming Language :: Python :: 3.2',
'Programming Language :: Python :: 3.3',
'Programming Language :: Python :: 3.4',
'Programming Language :: Python :: 3.5',
'Programming Language :: Python :: 3.6',
'Programming Language :: Python :: 3.7', 'Topic :: Utilities'
], )
shutil.rmtree('ppocr')
shutil.rmtree('tools')
shutil.rmtree('ppstructure')
os.remove('LICENSE')
ppstructure/table/README.md
View file @ 5d35c733
# Table structure and content prediction # Table structure
## 1. pipeline ## 1. pipeline
The ocr of the table mainly contains three models The ocr of the table mainly contains three models
...@@ -17,8 +17,26 @@ The table ocr flow chart is as follows ...@@ -17,8 +17,26 @@ The table ocr flow chart is as follows
## 2. How to use ## 2. How to use
### 2.1 quick start
### 2.1 Train ```python
cd PaddleOCR/ppstructure
# download model
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 table structure model of the ultra-lightweight Chinese OCR model and uncompress 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 ..
python3 table/predict_table.py --det_model_dir=inference/ch_ppocr_mobile_v2.0_det_infer --rec_model_dir=inference/ch_ppocr_mobile_v2.0_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
```
After running, the excel sheet of each picture will be saved in the directory specified by the output field
### 2.2 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 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
...@@ -48,7 +66,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo ...@@ -48,7 +66,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo
**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. **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.
### 2.2 Eval ### 2.3 Eval
The table uses TEDS (Tree-Edit-Distance-based Similarity) 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: The table uses TEDS (Tree-Edit-Distance-based Similarity) 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 ```json
...@@ -70,7 +88,7 @@ python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_di ...@@ -70,7 +88,7 @@ python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_di
``` ```
### 2.3 Inference ### 2.4 Inference
```python ```python
cd PaddleOCR/ppstructure cd PaddleOCR/ppstructure
......
ppstructure/table/README_ch.md
View file @ 5d35c733
# Table OCR # 表格结构化
## 1. Table OCR pineline ## 1. 表格结构化 pineline
表格的ocr主要包含三个模型 表格的ocr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
...@@ -19,7 +19,26 @@ ...@@ -19,7 +19,26 @@
## 2. 使用 ## 2. 使用
### 2.1 训练 ### 2.1 快速开始
```python
cd PaddleOCR/ppstructure
# 下载模型
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/ch_ppocr_mobile_v2.0_det_infer --rec_model_dir=inference/ch_ppocr_mobile_v2.0_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
### 2.2 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。 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。
#### 数据准备 #### 数据准备
...@@ -46,7 +65,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo ...@@ -46,7 +65,7 @@ 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 ### 2.3 评估
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) 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
...@@ -56,7 +75,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo ...@@ -56,7 +75,7 @@ python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./yo
[["<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"]] [["<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为对应的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)
...@@ -67,10 +86,9 @@ cd PaddleOCR/ppstructure ...@@ -67,10 +86,9 @@ 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
``` ```
### 2.4 预测
### 2.3 预测
```python ```python
cd PaddleOCR/ppstructure 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
``` ```
\ No newline at end of file
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
ppstructure/utility.py
View file @ 5d35c733
...@@ -36,7 +36,7 @@ def parse_args(): ...@@ -36,7 +36,7 @@ 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 = [], [], []
......
setup.py
View file @ 5d35c733
...@@ -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',
......
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