直接上代码：

from selenium import webdriver
from selenium.webdriver.common.action_chains import ActionChains
import time,re,requests
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from PIL import Image

import os,cv2
import sys
path = os.path.dirname(os.path.dirname(__file__))
sys.path.append(path)

class main():
    def __init__(self):
        self.url = 'https://static-mp-dc3bab1b-06be-41ca-9070-ab7368c17ae5.next.bspapp.com/'
        self.distance = 0
        self.left = 0
        self.track = []

    # 启动浏览器
    def Launch_browser(self):
        options = webdriver.ChromeOptions()
        options.add_argument('--headless')
        # self.driver = webdriver.Chrome(options=options)
        self.driver = webdriver.Chrome()
        self.wait = WebDriverWait(self.driver, 10, 0.5)
        self.driver.get(self.url)
        self.driver.find_element(By.XPATH,'/html/body/button').click()

        # 等待className为geetest_slider_button的元素在元素表中出现
        time.sleep(5)
        element = WebDriverWait(self.driver, 10).until(
            EC.visibility_of_element_located((By.CLASS_NAME, 'tcaptcha-transform'))
        )
        time.sleep(5)
        # 切换到iframe
        # 假设iframe有id或者其他属性，可以通过这些属性定位
        self.iframe = self.driver.find_element(By.ID,'tcaptcha_iframe_dy')
        self.driver.switch_to.frame(self.iframe)
        self.slider = self.driver.find_element(By.XPATH, '/html/body/div/div[3]/div[2]/div[7]')
        self.sliderImg = self.driver.find_element(By.XPATH, '/html/body/div/div[3]/div[2]/div[1]/div[2]/div')
        sliderImg_background_image_url = self.sliderImg.value_of_css_property('background-image')
        sliderImg_background_image_url = sliderImg_background_image_url[5:len(sliderImg_background_image_url) - 3]
        resp = requests.get(sliderImg_background_image_url)
        with open('./sliderImg.png', 'wb') as f:
            f.write(resp.content)
        slider_background_image_url = self.slider.value_of_css_property('background-image')
        slider_background_image_url = slider_background_image_url[5:len(slider_background_image_url) - 3]
        resp = requests.get(slider_background_image_url)
        with open('./slider.png', 'wb') as f:
            f.write(resp.content)
        # 150,270
        # 500,600
        image = Image.open('./slider.png')
        bg = image.crop([130, 479, 272, 622])
        bg.save('slider.png')
        import ddddocr
        det = ddddocr.DdddOcr(det=False, ocr=True, show_ad=False)
        with open('slider.png', 'rb') as f:
            target_bytes = f.read()
        with open('sliderImg.png', 'rb') as f:
            background_bytes = f.read()

        res = det.slide_match(target_bytes, background_bytes, simple_target=True)
        print(res)
        self.distance = res['target'][0]
        self.left = self.slider.value_of_css_property('left').split('px')[0]
        self.left = eval(self.left)
        xoffset = int(self.distance * 0.51)
        print(xoffset)
        verify_img = cv2.imread('sliderImg.png')
        # 调用函数，得到x坐标
        x = get_pos(verify_img)
        x = int(x * 0.51) - 30

        # 实现拖拽滑动
        ActionChains(self.driver).click_and_hold(self.slider).perform()
        ActionChains(self.driver).move_by_offset(x, 0).perform()
        ActionChains(self.driver).release().perform()
        self.quit()

    # 关闭浏览器
    def quit(self):
        time.sleep(10)
        self.driver.quit()



    # main方法
    def main(self):
        self.Launch_browser()

        # self.cjy()
        # self.move()
        # self.quit()


# 定义一个处理图片缺口的函数，最后是返回x坐标，滑块移动不需要y坐标
def get_pos(image):
    # 首先使用高斯模糊去噪，噪声会影响边缘检测的准确性，因此首先要将噪声过滤掉
    blurred = cv2.GaussianBlur(image, (5, 5), 0, 0)

    # 边缘检测，得到图片轮廓
    canny = cv2.Canny(blurred, 200, 400)  # 200为最小阈值，400为最大阈值，可以修改阈值达到不同的效果

    # 轮廓检测
    # cv2.findContours()函数接受的参数为二值图，即黑白的（不是灰度图），所以读取的图像要先转成灰度的，再转成二值图，此处canny已经是二值图
    # contours：所有的轮廓像素坐标数组，hierarchy 轮廓之间的层次关系
    contours, hierarchy = cv2.findContours(canny, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    # print(contours, hierarchy)

    for i, contour in enumerate(contours):  # 对所有轮廓进行遍历
        M = cv2.moments(contour)  # 并计算每一个轮廓的力矩(Moment)，就可以得出物体的质心位置
        # print(M)
        if M['m00'] == 0:
            cx = cy = 0
        else:
            # 得到质心位置，打印这个轮廓的面积和周长，用于过滤
            cx, cy = M['m10'] / M['m00'], M['m01'] / M['m00']
            print(cv2.contourArea(contour), cv2.arcLength(contour, True))

        # 判断这个轮廓是否在这个面积和周长的范围内
        if 5000 < cv2.contourArea(contour) < 8000 and 300 < cv2.arcLength(contour, True) < 500:
            print(cx)
            if cx < 300:
                continue
            print(cv2.contourArea(contour))
            print(cv2.arcLength(contour, True))

            # 外接矩形，x，y是矩阵左上点的坐标，w，h是矩阵的宽和高
            x, y, w, h = cv2.boundingRect(contour)

            cv2.rectangle(image, (x, y), (x + w, y + h), (0, 0, 255), 2)  # 画出矩行
            # cv2.imshow('image', image)
            cv2.imwrite('111.jpg', image)  # 保存
            return x
    return 0
if __name__ == '__main__':
    ma = main()
    ma.main()

效果展示：

Tip：用的是腾讯提供的web端接入示例