验证码识别【1】--字母识别

---

前言

记录字母验证码–>古诗文网selenium自动登录
本文是通过python，selenium，ddddocr去完成
附源码（附解析)

一、源码

from selenium import webdriver  # # 驱动浏览器
from selenium.webdriver.common.by import By  # 选择器
from selenium.webdriver.common.keys import Keys  # 按键
from selenium.webdriver.support.wait import WebDriverWait  # 等待页面加载完毕，寻找某些元素
from selenium.webdriver.support import expected_conditions as EC  ##等待指定标签加载完毕0
import time
import ddddocr  # 验证码识别模块


class GS():
    def __init__(self):
        self.url = 'https://so.gushiwen.cn/user/login.aspx?from=http://so.gushiwen.cn/user/collect.aspx'
        self.options = webdriver.ChromeOptions()  # 配置文件对象
        self.options.add_experimental_option('excludeSwitches', ['enable-automation'])  # 设置无头模式
        self.browser = webdriver.Chrome(options=self.options)
    def code(self):
        self.browser.get(self.url)  # 打开古诗文登录网页
        wait = WebDriverWait(self.browser, 280)  # 等待事件
        wait.until(EC.presence_of_element_located((By.ID, 'email')))
        # 输入账号
        login_input1 = self.browser.find_element(By.ID, 'email')
        login_input1.send_keys('id')
        # 输入密码
        login_input2 = self.browser.find_element(By.ID, 'pwd')
        login_input2.send_keys('password')
        # 等待2秒
        time.sleep(2)
        # 验证码截图
        img = self.browser.find_element(By.CSS_SELECTOR, '#imgCode')
        # img.screenshot_as_png
        i_c = img.screenshot_as_png
        # 保存图片
        with open('gushiwen.png', 'wb') as f:
            f.write(i_c)
        # 通过ddddocr分析数据
        ocr = ddddocr.DdddOcr(show_ad=False)  # ocr实例对象  # show_ad=False 关闭广告
        code_text = ocr.classification(img.screenshot_as_png)  # ddddocr识别数据
        print(code_text)
        code_input = self.browser.find_element(By.XPATH, '//*[@id="code"]')
        time.sleep(0.5)
        code_input.send_keys(code_text)
        time.sleep(0.5)
        code_input.send_keys(Keys.ENTER)  # 回车
if __name__ == '__main__':
    gs = GS()
    gs.code()

总结

记录验证码识别