验证码识别

# encoding=utf-8
import sys
sys.path.append('/home/henson/Documents/coding/bill/captcha_recognition')
from  captcha_recognition.recognition_img import distinguish_captcha
from PIL import Image,ImageDraw,ImageChops
from selenium.common.exceptions import NoSuchElementException
from  PIL import Image
import selenium
import sys
import time
import re
import csv
import pytesser
# 打开验证码界面

import os
from selenium import webdriver
from selenium.webdriver.support.ui import WebDriverWait
headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:23.0) Gecko/20100101 Firefox/23.0'}
chromedriver = "/home/henson/Documents/pycharm/webdriver/chromedriver"
os.environ["webdriver.chrome.driver"] = chromedriver
driver = webdriver.Chrome(chromedriver)
driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")



"""
driver.get('http://www******')
    #driver.get(url)


for i in range(1,1000):
    driver.get_screenshot_as_file('screenshot.png')

    # 获取指定元素位置
    element = driver.find_element_by_id('imgCode')
    left = int(element.location['x'])
    top = int(element.location['y'])
    right = int(element.location['x'] + element.size['width'])
    bottom = int(element.location['y'] + element.size['height'])

    # 通过Image处理图像
    im = Image.open('screenshot.png')
    im = im.crop((left, top, right, bottom))
    im.save('/home/henson/Documents/coding/bill/code/code+('+str(i)+').png')
    driver.find_element_by_xpath('//*[@id="imgCode"]').click()  # 切换验证码

"""


def get_img(url):
    try:
        # driver.get('*****')
        driver.get(url)
        # 获取截图

        url_core = driver.current_url
        current_url = url_core
        while (current_url == url_core):
            driver.get_screenshot_as_file('screenshot.png')

            # 获取指定元素位置
            element = driver.find_element_by_id('imgCode')
            left = int(element.location['x'])
            top = int(element.location['y'])
            right = int(element.location['x'] + element.size['width'])
            bottom = int(element.location['y'] + element.size['height'])

            # 通过Image处理图像
            im = Image.open('screenshot.png')
            im = im.crop((left, top, right, bottom))
            im.save('code.png')

            inputElement = driver.find_element_by_name("txtCode")
            #get_img(url_core)  # 获取验证码的图片
            image = Image.open('code.png')
            code = distinguish_captcha(image)  # 返回识别结果
            print(code)
            inputElement.send_keys(code)
            driver.find_element_by_xpath('//*[@id="btnOk"]').click()
            try:
                alert = driver.switch_to_alert()
                time.sleep(2)
                print(alert.text)  # 打印警告对话框内容
                alert.accept()  # 关闭弹出的窗口
                driver.find_element_by_name("txtCode").clear()

            except Exception:
                pass
            current_url = driver.current_url
        #driver.close()

    except Exception:
        print(Exception)

1.先获取验证码的图片，此处用的是selenium+driver 截屏功能，再对获得的图片进行切割
2.导入了人家的captcha_recognition，用来做验证码的识别
captcha_recognition包里，对有验证码进行二值话，切割….不过最好根据自己的验证码进行收集，切割样本手动分类，提高识别度。

=======================>>>>>>
实践感觉效果算是不错的了，可以一用