PaddleOCR光学字符识别

OCR光学字符识别，将不同类型的文档，图像中的字符保存为可编辑和可搜索的数据。

PaddleOCR

PaddleOCR旨在打造一套丰富、领先、且实用的OCR工具库，助力使用者训练出更好的模型，并应用落地。

PPOCR系列高质量预训练模型，准确的识别效果
- 超轻量ppocr_mobile移动端系列：检测（3.0M）+方向分类器（1.4M）+ 识别（5.0M）= 9.4M
- 通用ppocr_server系列：检测（47.1M）+方向分类器（1.4M）+ 识别（94.9M）= 143.4M
- 支持中英文数字组合识别、竖排文本识别、长文本识别
- 支持80+多语言识别，详见多语言模型
丰富易用的OCR相关工具组件
- 半自动数据标注工具PPOCRLabel：支持快速高效的数据标注
- 数据合成工具Style-Text：批量合成大量与目标场景类似的图像
支持用户自定义训练，提供丰富的预测推理部署方案
支持PIP快速安装使用
可运行于Linux、Windows、MacOS等多种系统
预安装 pytorch，安装过程参考ubuntu_NVIDIA30xx，开发环境显卡安装

安装paddleocr

在经历上述过程后，很容易安装paddleocr，运行以下命令：

pip install paddleocr

python -m pip install paddlepaddle-gpu==2.1.3.post112 -f https://www.paddlepaddle.org.cn/whl/windows/mkl/avx/stable.html

安装报错无法解决

安装gpu版本

pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple

安装cpu版本

pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple

输出结果

Looking in indexes: https://mirror.baidu.com/pypi/simple
Collecting paddlepaddle-gpu
  Downloading https://mirror.baidu.com/pypi/packages/ad/3f/074aa18eb969f452ebd712809756702fa7ad04aeea25960b77ed19c83870/paddlepaddle_gpu-2.1.3-cp38-cp38-win_amd64.whl (462.5 MB)
     |████████████████████████████████| 462.5 MB 1.9 kB/s
  Downloading https://mirror.baidu.com/pypi/packages/a4/23/13d2991c156cfd22bfd4a9ae6dcb1a9372004a0e16508b680d17f3280eb4/numpy-1.19.3-cp38-cp38-win_amd64.whl (13.3 MB)
     |████████████████████████████████| 13.3 MB 57 kB/s
Requirement already satisfied: certifi>=2017.4.17 in d:\users\vadmin\anaconda3\lib\site-packages (from requests>=2.20.0->paddlepaddle-gpu) (2020.12.5)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in d:\users\vadmin\anaconda3\lib\site-packages (from requests>=2.20.0->paddlepaddle-gpu) (1.26.4)
Requirement already satisfied: chardet<5,>=3.0.2 in d:\users\vadmin\anaconda3\lib\site-packages (from requests>=2.20.0->paddlepaddle-gpu) (4.0.0)
Requirement already satisfied: idna<3,>=2.5 in d:\users\vadmin\anaconda3\lib\site-packages (from requests>=2.20.0->paddlepaddle-gpu) (2.10)
Installing collected packages: numpy, gast, astor, paddlepaddle-gpu
  Attempting uninstall: numpy
    Found existing installation: numpy 1.20.1
    Uninstalling numpy-1.20.1:
      Successfully uninstalled numpy-1.20.1
Successfully installed astor-0.8.1 gast-0.3.3 numpy-1.19.3 paddlepaddle-gpu-2.1.3

https://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely
安装对应3.8的levenshtein、shapely

使用paddleocr

URL图片
Image='https://www.codekp.cn/download/img/ubuntu/NVIDIA1.png'
本地图片
Image='labelimg.jpg'

result = ocr.ocr(img_path)

结果

参考代码

from paddleocr import PaddleOCR, draw_ocr
from PIL import Image
# Paddleocr supports Chinese, English, French, German, Korean and Japanese.
# You can set the parameter `lang` as `ch`, `en`, `french`, `german`, `korean`, `japan`
# to switch the language model in order.
#ocr = PaddleOCR(use_angle_cls=True, lang='ch')  # need to run only once to download and load model into memory
ocr = PaddleOCR(lang='ch',use_gpu=False)
img_path = 'NVIDIA1.png'
result = ocr.ocr(img_path)
print(result)
for line in result:
    print(line)
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores)
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')