只需5行 Python 代码就可以轻松实现验证码识别，太稳了

pip install ddddocr

import ddddocr

ocr = ddddocr.DdddOcr()
with open('1.png', 'rb') as f:
    img_bytes = f.read()
res = ocr.classification(img_bytes)

print(res)

这个是最快的方法

当然还有其他方法

url = 'https://test.com/login'
driver.get(url)

获取登录页面需要输入账号密码进行模拟登录操作

user = driver.find_element(By.XPATH,'//*[@id="login"]/div[2]/div/form[2]/div[2]/div/div/input').send_keys('username')
pwd = driver.find_element(By.XPATH,'//*[@id="login"]/div[2]/div/form[2]/div[3]/div/div/input').send_keys('123456')

输入验证码

1.截图获取验证码图片

driver.find_element(By.XPATH,'//*[@id="login"]/div[2]/div/form[2]/div[4]/div/div/img').screenshot('verify.png')

2.将验证码图片发送给解码平台进行处理

解码平台：https://www.chaojiying.com/

第三方接口地址：

http://upload.chaojiying.net/Upload/Processing.php

url_code = 'http://upload.chaojiying.net/Upload/Processing.php'
data = {
# 用户名
'user': 'user',
# 密码
'pass': '123456',
# 用户的个人id，注册完账号之后并且登录，然后通过个人中心申请
'sofid': '123456',
# 验证码的类型
'codetype': 1004
}
files = {"userfile": open("verify.png", "rb")}

resp = requests.post(url_code, data=data, files=files)
res = resp.json()
print(res)
if res['err_no'] == 0:
    code = res['pic_str']
    print(f"识别成功，验证码为：{code}")
else:
    print("识别失败")

3.将解码后的验证码填写入验证码框中

verify = driver.find_element(By.XPATH,'//*[@id="login"]/div[2]/div/form[2]/div[4]/div/div/div/input').send_keys(code)

button = driver.find_element(By.XPATH,'//*[@id="login"]/div[2]/div/form[2]/div[5]/button').click()
sleep(5)

# 校验是否登录成功
url1 = 'https://test.com/#/serviceExpireManage'
driver.get(url1)

sleep(10)
driver.quit()