How to Automate 12306 Login with Selenium and Image Recognition (Full Code)

Introduction

This article demonstrates how to automate the login process of China Railway's 12306 website using Python, Selenium, and an image‑recognition service.

Tech Stack

Python, Selenium, image verification (Chaojiying), and handling of the slider captcha.

Approach

First the script captures the captcha image, sends it to the Chaojiying API to obtain click coordinates, then uses Selenium's ActionChains to click the points. After entering username and password, it modifies Selenium's navigator.webdriver property to bypass detection, and finally solves the slider captcha by dragging the slider, retrying if necessary.

Key Challenges

Image captcha recognition using an external service.

Bypassing 12306's anti‑automation measures for the slider captcha.

Code

from selenium import webdriver
from hashlib import md5
import requests
import time
from selenium.webdriver import ActionChains

# 这个类是超级鹰平台写的调用服务的接口代码，也是比较容易看懂的
class Chaojiying_Client(object):
    def __init__(self, username, password, soft_id):
        self.username = username
        password = password.encode('utf8')
        self.password = md5(password).hexdigest()
        self.soft_id = soft_id
        self.base_params = {
            'user': self.username,
            'pass2': self.password,
            'softid': self.soft_id,
        }
        self.headers = {
            'Connection': 'Keep-Alive',
            'User-Agent': 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0)'
        }

    def PostPic(self, im, codetype):
        """im: 图片字节
        codetype: 题目类型 参考 http://www.chaojiying.com/price.html"""
        params = {'codetype': codetype}
        params.update(self.base_params)
        files = {'userfile': ('ccc.jpg', im)}
        r = requests.post('http://upload.chaojiying.net/Upload/Processing.php', data=params, files=files, headers=self.headers)
        return r.json()

    def ReportError(self, im_id):
        """im_id:报错题目的图片ID"""
        params = {'id': im_id}
        params.update(self.base_params)
        r = requests.post('http://upload.chaojiying.net/Upload/ReportError.php', data=params, headers=self.headers)
        return r.json()

# 实例化浏览器，并且最大化。然后请求12306主网站，我这里是从首页请求的，大家可以直接从登陆页面请求
browser = webdriver.Chrome()
browser.maximize_window()
browser.get('http://12306.cn/')
time.sleep(5)
# 因为是从首页请求的，所以下面有两个点击的动作，都是为了点进登陆页面
browser.find_element_by_xpath('//*[@id="J-header-login"]/a[1]').click()
time.sleep(0.3)
# 这里比较重要了，这里就是利用这个代码，来更改selenium中的滑动功能，让网站不报错
script = 'Object.defineProperty(navigator,"webdriver",{get:()=>undefined,});'
browser.execute_script(script)
time.sleep(1)
# 这里进入帐号登录
browser.find_element_by_xpath('/html/body/div[2]/div[2]/ul/li[2]/a').click()
time.sleep(0.3)
# 这里直接定位验证码的png文件，然后保存
img = browser.find_element_by_xpath('//*[@id="J-loginImg"]')
img.screenshot('cde.png')
# 调用超级鹰的参数
chaojiying = Chaojiying_Client('用户名', '密码', 'ID')
im = open('../12306/cde.png', 'rb').read()
# 注意，这里返回的是一个字典格式，所以直接取要用的key，来返回坐标
result = chaojiying.PostPic(im, 9004)['pic_str']
print(result)
# 这里就是处理超级鹰返回坐标的方法了
all_list = []
# 通过判断超级鹰返回坐标的格式进行坐标处理，返回的坐标有两种形式，一种是以|隔开的，一种是用,隔开的，对应下面两种处理方式
# 处理好的坐标存入list
if '|' in result:
    list = result.split('|')
    for i in range(len(list)):
        x_y = []
        x = int(list[i].split(',')[0])
        y = int(list[i].split(',')[1])
        x_y.append(x)
        x_y.append(y)
        all_list.append(x_y)
else:
    x_y = []
    x = int(result.split(',')[0])
    y = int(result.split(',')[1])
    x_y.append(x)
    x_y.append(y)
    all_list.append(x_y)
print(all_list)
# 处理好的坐标进行循环，并带入selenium进行点击点击
for l in all_list:
    x = l[0]
    y = l[1]
    ActionChains(browser).move_to_element_with_offset(img, x, y).click().perform()
    time.sleep(0.5)
# 图片点击好以后，向表单内发送账户密码
browser.find_element_by_xpath('//*[@id="J-userName"]').send_keys('账号')
browser.find_element_by_xpath('//*[@id="J-password"]').send_keys('密码')
# 进行点击登录按钮
browser.find_element_by_xpath('//*[@id="J-login"]').click()
time.sleep(2)
# 下面就是滑动模块了
# 上面已经更改过selenium的滑动模块，所以这里就可以直接定位到按钮的位置，进行点击滑动
span = browser.find_element_by_xpath('//*[@id="nc_1_n1z"]')
action = ActionChains(browser)
# 这里是selenium的方法，按住点击不放
action.click_and_hold(span)
# 下面就是滑动了
action.drag_and_drop_by_offset(span, 400, 0).perform()
# 这里加了个循环，就是滑动不行，一直刷新继续滑动，直到成功
while True:
    try:
        info = browser.find_element_by_xpath('//*[@id="J-slide-passcode"]/div/span').text
        print(info)
        if info == '哎呀，出错了，点击刷新再来一次':
            # 点击刷新
            browser.find_element_by_xpath('//*[@id="J-slide-passcode"]/div/span/a').click()
            time.sleep(0.2)
            # 重新移动滑块
            span = browser.find_element_by_xpath('//*[@id="nc_1_n1z"]')
            action = ActionChains(browser)
            action.click_and_hold(span).perform()
            action.drag_and_drop_by_offset(span, 400, 0).perform()
            time.sleep(5)
    except:
        print('ok!')
        break
# 完成后，松开鼠标
action.release()
time.sleep(5)
# 退出
browser.quit()

Conclusion

The script successfully logs into 12306, but further development is required to implement ticket‑purchasing automation.