Python结合Selenium与Pillow实现图片验证码自动识别登录

裁剪出验证码图片

#coding=utf-8
from selenium import webdriver
import time
from PIL import Image
from selenium.webdriver.support.wait import WebDriverWait
driver = webdriver.Chrome()
# 进入该网站
driver.get("http://www2.nmec.org.cn/wangbao/nme/sp/root/account/signup.html")
# 能否在5s内找到验证码元素，能才继续
if WebDriverWait(driver,5).until(lambda the_driver:the_driver.find_element_by_id("CaptchaImg"), "查找不到该元素"):
 # 对于一次截屏无法到截到验证码的情况，需要滚动一段距离，然后验证码的y坐标也应该减去这段距离
 scroll = 500
 js = "document.documentElement.scrollTop='%s'" %scroll
 driver.execute_script(js)
 # 截下该网站的图片
 driver.get_screenshot_as_file("E:/Python_selenium_advance/Picture/full.png")
 # 获得这个图片元素
 img_ele = driver.find_element_by_id("CaptchaImg")
 # 得到该元素左上角的 x，y 坐标和右下角的 x，y 坐标
 left = img_ele.location.get('x')
 upper = img_ele.location.get('y') - 500
 right = left + img_ele.size.get('width')
 lower = upper + img_ele.size.get('height')
 # 打开之前的截图
 img = Image.open("E:/Python_selenium_advance/Picture/full.png")
 # 对截图进行裁剪，裁剪的范围为之前验证的左上角至右下角范围
 new_img = img.crop((left, upper, right, lower))
 # 裁剪完成之后保存到指定路径
 new_img.save("E:/Python_selenium_advance/Picture/croped.png")
 
 time.sleep(2)
 driver.quit()
else:
 print("找不到验证码元素")

使用图鉴商用接口识别验证码

import json
import requests
import base64
from io import BytesIO
from PIL import Image
from sys import version_info

def base64_api(uname, pwd, softid, img):
 img = img.convert('RGB')
 buffered = BytesIO()
 img.save(buffered, format="JPEG")
 if version_info.major >= 3:
  b64 = str(base64.b64encode(buffered.getvalue()), encoding='utf-8')
 else:
  b64 = str(base64.b64encode(buffered.getvalue()))
 data = {"username": uname, "password": pwd, "softid": softid, "image": b64}
 result = json.loads(requests.post("http://api.ttshitu.com/base64", json=data).text)
 if result['success']:
  return result["data"]["result"]
 else:
  return result["message"]
 return ""

from analysis_captcha import base64_api

def analysis_captcha(filename):
 '''
 使用 图鉴 商用接口来识别指定位置的验证码图片
 :param filename: 验证码图片位置
 :return : 验证码文本
 '''
 img_path = filename
 img = Image.open(img_path)
 result = base64_api(uname='kaibin', pwd='******', softid='4545454', img=img)
 return result