Python OCR 识别图片内容

1. 调用百度文字识别接口&识别图片内容

百度识别接口申请地址

修改MyPath 路径即可

MyPath = 'E:\\pdf-图片\\'
filesoure = MyPath


def baiduduqu(filesoure, filename):
    from aip import AipOcr
    import re
    import os
    APP_ID = '26646243'
    API_KEY = 'uH8dMfFHRBI0DTr6yzGCoD4X'
    SECRET_KEY = ''
    client = AipOcr(APP_ID, API_KEY, SECRET_KEY)
    dakai = open(filename, 'rb')
    duqu = dakai.read()
    message = client.basicGeneral(duqu)
    result = message.get('words_result')
    if result is not None:
        for duqu in message.get('words_result'):
            print(duqu.get('words'))
            with open("1pdf.txt", encoding='utf-8', mode="a") as f:
                f.write(duqu.get('words') + '\n')


def filename(fielsoure, filetype):
    import os
    pathDir = os.listdir(filesoure)
    for allDir in pathDir:
        child = os.path.join('%s%s' % (filesoure, allDir))
        print(child)
        baiduduqu(filesoure, child)


def run():
    import os
    os.chdir(filesoure)
    for i in os.listdir(os.getcwd()):
        postfix = os.path.splitext(i)[1]
        if postfix == '.jpeg' or postfix == '.png':
            filename(filesoure, postfix)


if __name__ == '__main__':
    run()

2. 使用easyocr库识别

2.1 安装Python 3环境

Python官网

2.2 安装easyocr

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple/ easyocr

2.3 安装模型

网盘链接:

https://pan.baidu.com/s/1xyPsKORak447jS82Pt-JHg 

提取码: onvl
复制到用户目录（python安装路径lib里面搜索） .EasyOCR\model

2.4 安装cv2

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple/ opencv-python

2.5 代码

model_storage_directory 是模型所在本地路径

import easyocr


def run():
    reader = easyocr.Reader(['ch_sim', 'en'], gpu=False, model_storage_directory=r'C:\Users\Work\AppData\Local\Programs'
                                                                                 r'\Python\Python37\Lib\site-packages'
                                                                                 r'\easyocr\model')
    context = ''
    for i in range(53, 1507):
        print('#' * 200 + ' 第{}页'.format(i))
        path = r'E:\pdfImage-1\{}.png'.format(i)
        result = reader.readtext(path)
        print('#' * 200 + '长度：{}'.format(len(result)))
        for j in range(5, len(result)):
            print(result[j][1])


if __name__ == '__main__':
    run()