python识别验证码登陆学校网站

约定不等于承诺〃 2022-09-10 05:16 178阅读 0赞

**使用技术**  
网络获取requests  
网络解析bs4  
验证码识别模块tesserocr

**登陆的网站**  
http://sso.sdwz.cn/cas/captcha

**代码**

import requests
    from bs4 import BeautifulSoup
    import urllib.request
    import os
    from urllib import parse
    import tesserocr
    from PIL import Image
    
    url = "http://sso.sdwz.cn/cas/login"
    
    s = requests.Session()
    
    
    
    
    def getImage():
        captcha = "http://sso.sdwz.cn/cas/captcha"
        img = s.get(captcha).content
        with open("1.jpg", "wb") as f:
            f.write(img)
    
        current_dir = os.path.dirname(os.path.abspath(__file__))
        image = Image.open(current_dir + "\\1.jpg")
        image = image.convert('L')  # 转化为灰度图
        threshold = 127  # 设定的二值化阈值
        table = []  # table是设定的一个表，下面的for循环可以理解为一个规则，小于阈值的，就设定为0，大于阈值的，就设定为1
        for i in range(256):
            if i < threshold:
                table.append(0)
            else:
                table.append(1)
    
        image = image.point(table, '1')  # 对灰度图进行二值化处理，按照table的规则（也就是上面的for循环）
        result = tesserocr.image_to_text(image)  # 对去噪后的图片进行识
        print("验证码为--", result)
        return result.replace("\n", "").replace(" ", "")
    
    
    def post(session, captcha, execution):
        global url
        headers = { 
            "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36",
            "Cookie": "SESSION=" + session,
            "Content-Type": "application/x-www-form-urlencoded",
            "Upgrade-Insecure-Requests": "1"
        }
        data = { 
            "execution": execution,
            "_eventId": "submit",
            "geolocation": "",
            "type": 1,
            "username": "xxxx",
            "password": "xxxx",
            "captcha": captcha,
            "phone": "",
            "smsCode": "",
            "_rememberMe": "on"
        }
        print(data)
    
        payload = parse.urlencode(data)
        print(payload)
        html = s.post(url, headers=headers, data=payload, verify=False)
        # 打印登录成功后的页面
        print(html.text)
    
    
    def get():
        global url
        headers = { 
            "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36",
        }
        data = s.get(url, headers=headers)
        cookies = data.cookies.get_dict()
        se = cookies["SESSION"]
        # flowKeyInfoHref
        subject = BeautifulSoup(data.text, "html.parser")
        execution = subject.select(".flowKeyInfoHref")[0]["href"]
        print("execution--", execution)
        print("session---", se)
        post(se, getImage(), execution)
    
    
    if __name__ == '__main__':
        get()

**使用selenium模拟登陆**

import requests
    
    import cv2 as cv
    import tesserocr
    from PIL import Image
    import urllib.request
    import os
    
    from selenium import webdriver
    from time import sleep
    
    # pip install selenium
    
    # 2017249042
    # 1074827413zhY
    
    # 获取二维码
    def getImage(browser):
        # 获取验证码的网址
        # url = "http://sso.sdwz.cn/cas/captcha"
        # rsp = urllib.request.urlopen(url)
        # img = rsp.read()
        # with open("1.jpg", "wb") as f:
        # f.write(img)
        browser.save_screenshot('1.png')
        imgelement = browser.find_element_by_class_name('code-img')
        location = imgelement.location
        print(location)
        size = imgelement.size
        print(size)
        # x = int(location['x'])
        # y = int(location['y'])
        x = 1100
        y = 520
        # width = int(size['width'])
        # height = int(size["height"])
        width = 120
        height = 50
        rangle = (x, y, x + width, y + height)
        print(rangle)
        i = Image.open("1.png")  # 打开截图
        result = i.crop(rangle)  # 使用Image的crop函数，从截图中再次截取我们需要的区域
        current_dir = os.path.dirname(os.path.abspath(__file__))
        result.save(current_dir + "\\result.png")
        image = Image.open(current_dir + "\\result.png")
    
        image = image.convert('L')  # 转化为灰度图
        threshold = 127  # 设定的二值化阈值
        table = []  # table是设定的一个表，下面的for循环可以理解为一个规则，小于阈值的，就设定为0，大于阈值的，就设定为1
        for i in range(256):
            if i < threshold:
                table.append(0)
            else:
                table.append(1)
    
        image = image.point(table, '1')  # 对灰度图进行二值化处理，按照table的规则（也就是上面的for循环）
        result = tesserocr.image_to_text(image)  # 对去噪后的图片进行识
    
        return result.replace(" ", "")
    
    
    # 进行登录
    def login():
        # http://npm.taobao.org/mirrors/chromedriver/
        # 谷歌的驱动器
        headers = { 
            "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36",
            "Content-Type": "application/x-www-form-urlencoded"
        }
        url = "http://sso.sdwz.cn/cas/login"
        data = { 
            "execution": "e2s1",
            "_eventId": "submit",
            "geolocation": "",
            "type": 1,
            "username": "2017249042",
            "password": "1074827413zhY",
            "captcha": "T1zg",
            "phone": "",
            "smsCode": "",
            "_rememberMe": "on"
        }
        #
        browser = webdriver.Chrome()
        browser.minimize_window()
        browser.get('http://sso.sdwz.cn/cas/login')
        # 设置用户名和密码
        browser.find_element_by_id("username").clear()
        browser.find_element_by_id("username").send_keys("2017249042")
        browser.find_element_by_id("password").clear()
        browser.find_element_by_id("password").send_keys("1074827413zhY")
        # 添加验证码
        result = getImage(browser)
        print("验证码",result)
        browser.find_element_by_id("captcha").clear()
        browser.find_element_by_id("captcha").send_keys(result)
        # 点击按钮
        browser.find_element_by_id("login-phone").click()
        # 打印页面
        print(browser.page_source)
        sleep(10)
        # 关闭
        browser.close()
    
    
    if __name__ == '__main__':
        # 下载安装tesseract-orc
        # https://digi.bib.uni-mannheim.de/tesseract/
        login()