Python基于内置库pytesseract实现图片验证码识别功能

　　pip install pillow #一个python的图像处理库，pytesseract依赖
　　pip install pytesseract

import requests
from PIL import Image
import pytesseract

'''
简单验证码
'''
# 验证码地址
url = "https://www.renrendai.com/passport/index/captcha?time=1551682134111"
response = requests.get(url).content

#将图片写入文件
with open('yzm.png','wb') as f:
  f.write(response)
f.close()

'''识别验证码'''
#第一步：通过内置模块PIL打开文件
pic = Image.open('yzm.png')
#第二步：识别图片中的内容
pic_str = pytesseract.image_to_string(pic)
print("验证码识别结果为:",pic_str)

#下载图片
baidu_url = "https://wkretype.bdimg.com/retype/zoom/4127ed79a26925c52cc5bf99?pn=2&o=jpg_6&md5sum=9cdc209bc34a40ed774f7e14c0be59c4&sign=5dbcb28bf1&png=11238-22475&jpg=41808-117940"
baidu_pic = requests.get(baidu_url).content

#图片写入文件
with open('baidu_pic.jpg','wb') as f:
  f.write(baidu_pic)
f.close()

#识别验证码
baidu_img = Image.open('baidu_pic.jpg')
baidu_img_str = pytesseract.image_to_string(baidu_img,lang="chi_sim")
print('百度文库图片内容为:',baidu_img_str)

from chaojiying import Chaojiying

chaojiying_url= "http://www.chaojiying.com/include/code/code.php?u=1"
response = requests.get(chaojiying_url).content

with open('rryz.png','wb') as f:
  f.write(response)
f.close()

#读取文件内容
with open('rryz.png','rb') as f:
  pic1 = f.read()

#调用第三方打码平台接口识别验证码
yz = Chaojiying(username='*****', password='****', soft_id='****')

res = yz.post_pic(pic1,codetype='1902').get('pic_str') #1902 验证码类型
print('识别的结果：',res)