Awesome OCR toolkits based on PaddlePaddle （8.6M ultra-lightweight pre-trained model, support training and deployment among server, mobile, embeded and IoT devices

Navigation

Verified details

These details have been verified by PyPI
Maintainers
Avatar for jzhang533 from gravatar.com jzhang533 Avatar for zhoujun from gravatar.com zhoujun

Unverified details

These details have not been verified by PyPI
Project links
GitHub Statistics

View statistics for this project via Libraries.io, or by using our public dataset on Google BigQuery

Meta

License: Apache License 2.0

Author: Baidu PaddlePaddle

Tags ocr, textdetection, textrecognition, paddleocr, crnn, east, star-net, rosetta, ocrlite, db, chineseocr, chinesetextdetection, chinesetextrecognition

Classifiers

Project description

ppcor package使用说明

快速上手

安装whl包

pip安装

pip install paddleocr

本地构建并安装

python setup.py bdist_wheel
pip install dist/paddleocr-0.0.1-py3-none-any.whl

1. 通过代码使用

检测+识别全流程

from paddleocr import PaddleOCR
ocr = PaddleOCR(model_storage_directory='./model') # need to run only once to load model into memory
img = 'PaddleOCR/doc/imgs/11.jpg'
result = ocr.ocr(img)
for line in result:
    print(line)

结果是一个list，每个item包含了文本框，文字和识别置信度

[[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]], ['纯臻营养护发素', 0.964739]]
[[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]], ['产品信息/参数', 0.98069626]]
[[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]], ['（45元/每公斤，100公斤起订）', 0.9676722]]
[[[22.0, 140.0], [284.0, 140.0], [284.0, 167.0], [22.0, 167.0]], ['每瓶22元，1000瓶起订）', 0.97444016]]
[[[22.0, 174.0], [85.0, 174.0], [85.0, 198.0], [22.0, 198.0]], ['【品牌】', 0.8187138]]
[[[89.0, 176.0], [301.0, 176.0], [301.0, 196.0], [89.0, 196.0]], ['：代加工方式/OEMODM', 0.9421848]]
[[[23.0, 205.0], [85.0, 205.0], [85.0, 229.0], [23.0, 229.0]], ['【品名】', 0.76008326]]
[[[88.0, 204.0], [235.0, 206.0], [235.0, 229.0], [88.0, 227.0]], ['：纯臻营养护发素', 0.9633639]]
[[[23.0, 236.0], [121.0, 236.0], [121.0, 261.0], [23.0, 261.0]], ['【产品编号】', 0.84101385]]
[[[110.0, 239.0], [239.0, 239.0], [239.0, 256.0], [110.0, 256.0]], ['1：YM-X-3011', 0.8621878]]
[[[414.0, 233.0], [430.0, 233.0], [430.0, 304.0], [414.0, 304.0]], ['ODM OEM', 0.9084018]]
[[[23.0, 268.0], [183.0, 268.0], [183.0, 292.0], [23.0, 292.0]], ['【净含量】：220ml', 0.9278281]]
[[[24.0, 301.0], [118.0, 301.0], [118.0, 321.0], [24.0, 321.0]], ['【适用人群】', 0.90901047]]
[[[127.0, 300.0], [254.0, 300.0], [254.0, 323.0], [127.0, 323.0]], ['：适合所有肤质', 0.95465785]]
[[[24.0, 332.0], [117.0, 332.0], [117.0, 353.0], [24.0, 353.0]], ['【主要成分】', 0.88936955]]
[[[139.0, 332.0], [236.0, 332.0], [236.0, 352.0], [139.0, 352.0]], ['鲸蜡硬脂醇', 0.9447544]]
[[[248.0, 332.0], [345.0, 332.0], [345.0, 352.0], [248.0, 352.0]], ['燕麦B-葡聚', 0.89748293]]
[[[54.0, 363.0], [232.0, 363.0], [232.0, 383.0], [54.0, 383.0]], [' 椰油酰胺丙基甜菜碱', 0.902023]]
[[[25.0, 364.0], [64.0, 364.0], [64.0, 383.0], [25.0, 383.0]], ['糖、', 0.985203]]
[[[244.0, 363.0], [281.0, 363.0], [281.0, 382.0], [244.0, 382.0]], ['泛服', 0.44537082]]
[[[367.0, 367.0], [475.0, 367.0], [475.0, 388.0], [367.0, 388.0]], ['（成品包材）', 0.9834532]]
[[[24.0, 395.0], [120.0, 395.0], [120.0, 416.0], [24.0, 416.0]], ['【主要功能】', 0.88684446]]
[[[128.0, 397.0], [273.0, 397.0], [273.0, 414.0], [128.0, 414.0]], ['：可紧致头发磷层', 0.9342501]]
[[[265.0, 395.0], [361.0, 395.0], [361.0, 415.0], [265.0, 415.0]], ['琴，从而达到', 0.8253762]]
[[[25.0, 425.0], [372.0, 425.0], [372.0, 448.0], [25.0, 448.0]], ['即时持久改善头发光泽的效果，给干燥的头', 0.97785276]]
[[[26.0, 457.0], [137.0, 457.0], [137.0, 477.0], [26.0, 477.0]], ['发足够的滋养', 0.9577897]]

单独执行检测

from paddleocr import PaddleOCR
ocr = PaddleOCR(model_storage_directory='./model') # need to run only once to load model into memory
img = 'PaddleOCR/doc/imgs/11.jpg'
result = ocr.ocr(img,rec=False)
for line in result:
    print(line)

结果是一个list，每个item只包含文本框

[[26.0, 457.0], [137.0, 457.0], [137.0, 477.0], [26.0, 477.0]]
[[25.0, 425.0], [372.0, 425.0], [372.0, 448.0], [25.0, 448.0]]
[[128.0, 397.0], [273.0, 397.0], [273.0, 414.0], [128.0, 414.0]]
[[265.0, 395.0], [361.0, 395.0], [361.0, 415.0], [265.0, 415.0]]
[[24.0, 395.0], [120.0, 395.0], [120.0, 416.0], [24.0, 416.0]]
[[367.0, 367.0], [475.0, 367.0], [475.0, 388.0], [367.0, 388.0]]
[[54.0, 363.0], [232.0, 363.0], [232.0, 383.0], [54.0, 383.0]]
[[25.0, 364.0], [64.0, 364.0], [64.0, 383.0], [25.0, 383.0]]
[[244.0, 363.0], [281.0, 363.0], [281.0, 382.0], [244.0, 382.0]]
[[248.0, 332.0], [345.0, 332.0], [345.0, 352.0], [248.0, 352.0]]
[[139.0, 332.0], [236.0, 332.0], [236.0, 352.0], [139.0, 352.0]]
[[24.0, 332.0], [117.0, 332.0], [117.0, 353.0], [24.0, 353.0]]
[[127.0, 300.0], [254.0, 300.0], [254.0, 323.0], [127.0, 323.0]]
[[24.0, 301.0], [118.0, 301.0], [118.0, 321.0], [24.0, 321.0]]
[[23.0, 268.0], [183.0, 268.0], [183.0, 292.0], [23.0, 292.0]]
[[110.0, 239.0], [239.0, 239.0], [239.0, 256.0], [110.0, 256.0]]
[[23.0, 236.0], [121.0, 236.0], [121.0, 261.0], [23.0, 261.0]]
[[414.0, 233.0], [430.0, 233.0], [430.0, 304.0], [414.0, 304.0]]
[[88.0, 204.0], [235.0, 206.0], [235.0, 229.0], [88.0, 227.0]]
[[23.0, 205.0], [85.0, 205.0], [85.0, 229.0], [23.0, 229.0]]
[[89.0, 176.0], [301.0, 176.0], [301.0, 196.0], [89.0, 196.0]]
[[22.0, 174.0], [85.0, 174.0], [85.0, 198.0], [22.0, 198.0]]
[[22.0, 140.0], [284.0, 140.0], [284.0, 167.0], [22.0, 167.0]]
[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]]
[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]]
[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]]

单独执行识别

from paddleocr import PaddleOCR
ocr = PaddleOCR(model_storage_directory='./model') # need to run only once to load model into memory
img = 'PaddleOCR/doc/imgs_words/ch/word_1.jpg'
result = ocr.ocr(img,det=False)
for line in result:
    print(line)

结果是一个list，每个item只包含识别结果和识别置信度

['韩国小馆', 0.9907421]

通过命令行使用

查看帮助信息

paddleocr -h

检测+识别全流程

paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg

结果是一个list，每个item包含了文本框，文字和识别置信度

[[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]], ['纯臻营养护发素', 0.964739]]
[[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]], ['产品信息/参数', 0.98069626]]
[[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]], ['（45元/每公斤，100公斤起订）', 0.9676722]]
[[[22.0, 140.0], [284.0, 140.0], [284.0, 167.0], [22.0, 167.0]], ['每瓶22元，1000瓶起订）', 0.97444016]]
[[[22.0, 174.0], [85.0, 174.0], [85.0, 198.0], [22.0, 198.0]], ['【品牌】', 0.8187138]]
[[[89.0, 176.0], [301.0, 176.0], [301.0, 196.0], [89.0, 196.0]], ['：代加工方式/OEMODM', 0.9421848]]
[[[23.0, 205.0], [85.0, 205.0], [85.0, 229.0], [23.0, 229.0]], ['【品名】', 0.76008326]]
[[[88.0, 204.0], [235.0, 206.0], [235.0, 229.0], [88.0, 227.0]], ['：纯臻营养护发素', 0.9633639]]
[[[23.0, 236.0], [121.0, 236.0], [121.0, 261.0], [23.0, 261.0]], ['【产品编号】', 0.84101385]]
[[[110.0, 239.0], [239.0, 239.0], [239.0, 256.0], [110.0, 256.0]], ['1：YM-X-3011', 0.8621878]]
[[[414.0, 233.0], [430.0, 233.0], [430.0, 304.0], [414.0, 304.0]], ['ODM OEM', 0.9084018]]
[[[23.0, 268.0], [183.0, 268.0], [183.0, 292.0], [23.0, 292.0]], ['【净含量】：220ml', 0.9278281]]
[[[24.0, 301.0], [118.0, 301.0], [118.0, 321.0], [24.0, 321.0]], ['【适用人群】', 0.90901047]]
[[[127.0, 300.0], [254.0, 300.0], [254.0, 323.0], [127.0, 323.0]], ['：适合所有肤质', 0.95465785]]
[[[24.0, 332.0], [117.0, 332.0], [117.0, 353.0], [24.0, 353.0]], ['【主要成分】', 0.88936955]]
[[[139.0, 332.0], [236.0, 332.0], [236.0, 352.0], [139.0, 352.0]], ['鲸蜡硬脂醇', 0.9447544]]
[[[248.0, 332.0], [345.0, 332.0], [345.0, 352.0], [248.0, 352.0]], ['燕麦B-葡聚', 0.89748293]]
[[[54.0, 363.0], [232.0, 363.0], [232.0, 383.0], [54.0, 383.0]], [' 椰油酰胺丙基甜菜碱', 0.902023]]
[[[25.0, 364.0], [64.0, 364.0], [64.0, 383.0], [25.0, 383.0]], ['糖、', 0.985203]]
[[[244.0, 363.0], [281.0, 363.0], [281.0, 382.0], [244.0, 382.0]], ['泛服', 0.44537082]]
[[[367.0, 367.0], [475.0, 367.0], [475.0, 388.0], [367.0, 388.0]], ['（成品包材）', 0.9834532]]
[[[24.0, 395.0], [120.0, 395.0], [120.0, 416.0], [24.0, 416.0]], ['【主要功能】', 0.88684446]]
[[[128.0, 397.0], [273.0, 397.0], [273.0, 414.0], [128.0, 414.0]], ['：可紧致头发磷层', 0.9342501]]
[[[265.0, 395.0], [361.0, 395.0], [361.0, 415.0], [265.0, 415.0]], ['琴，从而达到', 0.8253762]]
[[[25.0, 425.0], [372.0, 425.0], [372.0, 448.0], [25.0, 448.0]], ['即时持久改善头发光泽的效果，给干燥的头', 0.97785276]]
[[[26.0, 457.0], [137.0, 457.0], [137.0, 477.0], [26.0, 477.0]], ['发足够的滋养', 0.9577897]]

单独执行检测

paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --rec false

结果是一个list，每个item只包含文本框

[[26.0, 457.0], [137.0, 457.0], [137.0, 477.0], [26.0, 477.0]]
[[25.0, 425.0], [372.0, 425.0], [372.0, 448.0], [25.0, 448.0]]
[[128.0, 397.0], [273.0, 397.0], [273.0, 414.0], [128.0, 414.0]]
[[265.0, 395.0], [361.0, 395.0], [361.0, 415.0], [265.0, 415.0]]
[[24.0, 395.0], [120.0, 395.0], [120.0, 416.0], [24.0, 416.0]]
[[367.0, 367.0], [475.0, 367.0], [475.0, 388.0], [367.0, 388.0]]
[[54.0, 363.0], [232.0, 363.0], [232.0, 383.0], [54.0, 383.0]]
[[25.0, 364.0], [64.0, 364.0], [64.0, 383.0], [25.0, 383.0]]
[[244.0, 363.0], [281.0, 363.0], [281.0, 382.0], [244.0, 382.0]]
[[248.0, 332.0], [345.0, 332.0], [345.0, 352.0], [248.0, 352.0]]
[[139.0, 332.0], [236.0, 332.0], [236.0, 352.0], [139.0, 352.0]]
[[24.0, 332.0], [117.0, 332.0], [117.0, 353.0], [24.0, 353.0]]
[[127.0, 300.0], [254.0, 300.0], [254.0, 323.0], [127.0, 323.0]]
[[24.0, 301.0], [118.0, 301.0], [118.0, 321.0], [24.0, 321.0]]
[[23.0, 268.0], [183.0, 268.0], [183.0, 292.0], [23.0, 292.0]]
[[110.0, 239.0], [239.0, 239.0], [239.0, 256.0], [110.0, 256.0]]
[[23.0, 236.0], [121.0, 236.0], [121.0, 261.0], [23.0, 261.0]]
[[414.0, 233.0], [430.0, 233.0], [430.0, 304.0], [414.0, 304.0]]
[[88.0, 204.0], [235.0, 206.0], [235.0, 229.0], [88.0, 227.0]]
[[23.0, 205.0], [85.0, 205.0], [85.0, 229.0], [23.0, 229.0]]
[[89.0, 176.0], [301.0, 176.0], [301.0, 196.0], [89.0, 196.0]]
[[22.0, 174.0], [85.0, 174.0], [85.0, 198.0], [22.0, 198.0]]
[[22.0, 140.0], [284.0, 140.0], [284.0, 167.0], [22.0, 167.0]]
[[24.0, 109.0], [333.0, 109.0], [333.0, 136.0], [24.0, 136.0]]
[[24.0, 80.0], [172.0, 80.0], [172.0, 104.0], [24.0, 104.0]]
[[24.0, 36.0], [304.0, 34.0], [304.0, 72.0], [24.0, 74.0]]

单独执行识别

paddleocr --image_dir PaddleOCR/doc/imgs_words/ch/word_1.jpg --det false

结果是一个list，每个item只包含识别结果和识别置信度

['韩国小馆', 0.9907421]

参数说明

字段 说明 默认值
use_gpu 是否使用GPU TRUE
gpu_mem 初始化占用的GPU内存大小 8000M
image_dir 通过命令行调用时执行预测的图片或文件夹路径
det_algorithm 使用的检测算法类型 DB
det_model_name 有两种使用方式： 1. 检测算法名称，此名称必须在支持列表内(目前只内置了ch_det_mv3_db)，传入错误参数时会显示支持的列表 2. 自己转换好的inference模型路径，此时模型路径下必须包含model和params文件。选择此方式时，需要手动指定det_algorithm的值 ch_det_mv3_db
det_max_side_len 检测算法前向时图片长边的最大尺寸，当长边超出这个值时会将长边resize到这个大小，短边等比例缩放 960
det_db_thresh DB模型输出预测图的二值化阈值 0.3
det_db_box_thresh DB模型输出框的阈值，低于此值的预测框会被丢弃 0.5
det_db_unclip_ratio DB模型输出框扩大的比例 2
det_east_score_thresh EAST模型输出预测图的二值化阈值 0.8
det_east_cover_thresh EAST模型输出框的阈值，低于此值的预测框会被丢弃 0.1
det_east_nms_thresh EAST模型输出框NMS的阈值 0.2
rec_algorithm 使用的识别算法类型 CRNN
rec_model_name 有两种使用方式： 1. 识别算法名称，此名称必须在支持列表内(目前支持CRNN,Rosetta,STAR,RARE等算法,但是内置的只有ch_rec_mv3_crnn_enhance)，传入错误参数时会显示支持的列表 2. 自己转换好的inference模型路径，此时模型路径下必须包含model和params文件。选择此方式时，需要手动指定rec_algorithm的值 ch_rec_mv3_crnn_enhance
rec_image_shape 识别算法的输入图片尺寸 "3,32,320"
rec_char_type 识别算法的字符类型，中文(ch)或英文(en) ch
rec_batch_num 进行识别时，同时前向的图片数 30
rec_char_dict_path 识别模型字典路径，当rec_model_name使用方式2传参时需要修改为自己的路径
use_space_char 是否识别空格 TRUE
enable_mkldnn 是否启用mkldnn FALSE
model_storage_directory 下载模型保存路径 ～/.ppocr
det 前向时使用启动检测 TRUE
rec 前向时是否启动识别 TRUE

Project details

Verified details

These details have been verified by PyPI
Maintainers
Avatar for jzhang533 from gravatar.com jzhang533 Avatar for zhoujun from gravatar.com zhoujun

Unverified details

These details have not been verified by PyPI
Project links
GitHub Statistics

View statistics for this project via Libraries.io, or by using our public dataset on Google BigQuery

Meta

License: Apache License 2.0

Author: Baidu PaddlePaddle

Tags ocr, textdetection, textrecognition, paddleocr, crnn, east, star-net, rosetta, ocrlite, db, chineseocr, chinesetextdetection, chinesetextrecognition

Classifiers

Release history Release notifications | RSS feed

2.7.5 yanked

2.7.4 yanked

2.7.3

2.7.2

2.7.0.3

2.7.0.2

2.7.0.1

2.7.0.0

2.6.1.3

2.6.1.2

2.6.1.1

2.6.1.0

2.6.0.3

2.6.0.2

2.6.0.1

2.6

2.5.0.3

2.5.0.2

2.5

2.4.0.4

2.4.0.3

2.4.0.2

2.4.0.1

2.4

2.3.0.2

2.3.0.1

2.3

2.2.0.2

2.2.0.1

2.2

2.0.6

2.0.5

2.0.4

2.0.3

2.0.2

2.0.1

1.1.1

1.0.1

1.0.0

0.0.3.1

0.0.3

0.0.2

This version

0.0.1.1

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

paddleocr-0.0.1.1.tar.gz (209.7 kB view hashes)

Uploaded Source

Built Distribution

paddleocr-0.0.1.1-py3-none-any.whl (245.2 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page