【Python学习】图像识别-验证二维码

 安装

tesseract.exe 下载：https://digi.bib.uni-mannheim.de/tesseract/

chi_sim.traineddata：https://github.com/tesseract-ocr/tessdata/find/master/chi_sim.traineddata

 jTessBoxEditor2.0工具，用于调整图片上文字的内容和位置:https://www.jianshu.com/p/c8ba23ec672a

pip install pillow pytesseract

 测试代码

 1 # !/usr/bin/env python
 2 # coding:utf-8
 3 import requests
 4 # from lxml import etree
 5 import time
 6 import os
 7 import re
 8 import pytesseract
 9 from PIL import Image, ImageEnhance
10 
11 
12 def binarizing(img, threshold):  # #二值化。i参数 灰度图，阈值
13     pixdata = img.load()
14     w, h = img.size
15     for y in range(h):
16         for x in range(w):
17             if pixdata[x, y] < threshold:
18                 pixdata[x, y] = 0
19             else:
20                 pixdata[x, y] = 255
21     return img
22 
23 
24 def Compute_Correct_Rate(Correct, Recognition):  # 计算正确率，打印识别错误的
25     Correct_num = 0
26     for i in range(len(Correct)):
27         if Correct[i].lower() == Recognition[i].lower():
28             Correct_num += 1
29         else:
30             print('{} is Wrong,result is {}'.format(Recognition[i], Correct[i]))
31     print('the Correct Rate is ', Correct_num / len(Correct))
32 
33 
34 def process_img(rootPath):
35     Correct_result = []
36     Recognition_result = []
37     for root, dirs, files in os.walk(rootPath):  # 遍历整个文件夹，识别所有图片
38         for file in files:
39             filepath = os.path.join(rootPath, file)
40             img = Image.open(filepath)  # 打开图片文件
41             # imgry = img.convert('L')  # 转化为灰度图
42             # img = binarizing(imgry, 130)
43             img = img.convert('RGB')  # 这里也可以尝试使用L
44             enhancer = ImageEnhance.Color(img)
45             enhancer = enhancer.enhance(0)
46             enhancer = ImageEnhance.Brightness(enhancer)
47             enhancer = enhancer.enhance(2)
48             enhancer = ImageEnhance.Contrast(enhancer)
49             enhancer = enhancer.enhance(8)
50             enhancer = ImageEnhance.Sharpness(enhancer)
51             img = enhancer.enhance(20)
52             text = pytesseract.image_to_string(img, lang='eng')
53             match = re.findall(r'[\w]+', text)
54             if len(match) != 0:
55                 r_result = ''.join(match)
56             else:
57                 r_result = ''
58             c_result = os.path.splitext(file)[0]
59             Correct_result.append(c_result)
60             Recognition_result.append(r_result)
61 
62     Compute_Correct_Rate(Correct_result, Recognition_result)
63 
64 
65 if __name__ == '__main__':
66     process_img(r"D:\dxx\software\project\DevTools\MyTest\ImagePath")
View Code

 

遇到问题

1.没有安装tesseract.exe

 1 第一次用Pycharm使用tesseract的时候 需要修改一下pytesseract.py中的tesseract_cmd指向的路径
 2 c:\py368\Lib\site-packages\pytesseract\pytesseract.py文件中修改
 3 tesseract_cmd = ‘G:/soft/Tesseract-OCR验证码识别/tesseract’
 4 
 5 tesseract需要额外安装
 6 链接：https://pan.baidu.com/s/12kfe9bRCj8vFd5f9h5iPCw
 7 提取码：dutl
 8 复制这段内容后打开百度网盘手机App，操作更方便哦
 9 
10 PyCharm-执行程序PermissionError: [WinError 5] 拒绝访问
11 cmd命令需要精确到exe文件
12 
13 tesseract_cmd = ‘G:/soft/Tesseract-OCR验证码识别/tesseract.exe’
14 增加环境变量TESSDATA_PREFIX G:/soft/Tesseract-OCR验证码识别/tessdata
15 系统Path 变量增加G:/soft/Tesseract-OCR验证码识别/
16 打开命令终端，输入：tesseract -v，可以看到版本信息
17 用命令tesseract --list-langs来查看Tesseract-OCR支持语言
View Code

 

参考链接：

https://www.cnblogs.com/jclian91/p/9158372.html

https://www.jianshu.com/p/c8ba23ec672a

https://zhuanlan.zhihu.com/p/113961004

https://www.cnblogs.com/lizhe860/p/8969171.html

https://digi.bib.uni-mannheim.de/tesseract/

https://blog.csdn.net/m0_46498587/article/details/109255325

样例：https://www.pianshen.com/article/84901303763/

https://www.cnblogs.com/pythonywy/archive/2020/02/04/12258681.html

posted @ 2021-08-14 11:44  gtea  阅读(205)  评论(0)  编辑  收藏  举报