攻克滑动拼图反爬：Python 高效爬取网页图片实战案例

# 模拟浏览器操作
pip install selenium==4.15.0
# 图片处理与缺口识别
pip install opencv-python==4.8.0
# 网络请求与随机数生成
pip install requests numpy random

import cv2
import numpy as np
import random
import time
import requests
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options
import os

# 创建图片保存文件夹
if not os.path.exists("captcha_img"):
    os.mkdir("captcha_img")
if not os.path.exists("target_imgs"):
    os.mkdir("target_imgs")

def init_browser():
    """初始化Chrome浏览器，规避反爬检测"""
    chrome_options = Options()
    # 关闭自动化提示
    chrome_options.add_experimental_option("excludeSwitches", ["enable-automation"])
    chrome_options.add_experimental_option('useAutomationExtension', False)
    # 禁用图片加载（可选，提升速度）
    # chrome_options.add_argument('--blink-settings=imagesEnabled=false')
    # 初始化驱动
    driver = webdriver.Chrome(options=chrome_options)
    # 修改浏览器特征，隐藏webdriver属性
    driver.execute_cdp_cmd('Page.addScriptToEvaluateOnNewDocument', {
        'source': 'Object.defineProperty(navigator, "webdriver", {get: () => undefined})'
    })
    driver.maximize_window()
    return driver

# 初始化浏览器并访问目标网站
driver = init_browser()
# 替换为你要爬取的带滑动拼图验证的网站
target_url = "https://www.test-spider.com/slider-captcha"
driver.get(target_url)
time.sleep(2)

def get_gap_position(bg_path, slider_path):
    """识别缺口位置，返回横向偏移坐标"""
    # 读取图片
    bg = cv2.imread(bg_path, 0)
    slider = cv2.imread(slider_path, 0)
    
    # 模板匹配：查找滑块在背景图中的位置
    result = cv2.matchTemplate(bg, slider, cv2.TM_CCOEFF_NORMED)
    # 获取匹配结果的最大值坐标
    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(result)
    
    # 缺口横坐标（根据实际图片尺寸微调）
    gap_x = max_loc[0]
    print(f"识别到缺口横坐标：{gap_x}")
    return gap_x

# 获取缺口位置
gap_x = get_gap_position(bg_path, slider_path)

def get_slide_track(distance):
    """生成拟人滑动轨迹：返回每一步的偏移量"""
    track = []
    current = 0
    # 加速度与减速阈值
    mid = distance * 4 / 5
    t = 0.2
    v = 0
    
    while current < distance:
        if current < mid:
            a = 2  # 加速阶段
        else:
            a = -3  # 减速阶段
        v0 = v
        v = v0 + a * t
        move = v0 * t + 0.5 * a * t * t
        current += move
        # 添加微小抖动，更拟人
        track.append(round(move + random.uniform(-0.5, 0.5), 2))
    return track

def slide_captcha(driver, gap_x):
    """模拟滑动拼图验证"""
    # 定位滑块元素
    slider = driver.find_element(By.CLASS_NAME, "slider-btn")
    # 生成滑动轨迹
    track = get_slide_track(gap_x - 10)  # 微调偏移量
    
    # 按住滑块
    ActionChains(driver).click_and_hold(slider).perform()
    time.sleep(random.uniform(0.2, 0.5))
    
    # 按照轨迹滑动
    for step in track:
        ActionChains(driver).move_by_offset(xoffset=step, yoffset=0).perform()
        time.sleep(random.uniform(0.005, 0.01))
    
    # 随机停顿后释放
    time.sleep(random.uniform(0.3, 0.6))
    ActionChains(driver).release().perform()
    time.sleep(2)
    
    # 判断验证是否成功
    try:
        driver.find_element(By.CLASS_NAME, "captcha-success")
        print("滑动验证成功！")
        return True
    except:
        print("滑动验证失败，重试中...")
        return False

# 执行滑动验证，最多重试3次
retry_count = 0
while retry_count < 3:
    if slide_captcha(driver, gap_x):
        break
    retry_count += 1
    time.sleep(1)

def download_img(img_url, save_path):
    """下载单张图片"""
    try:
        headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
        }
        response = requests.get(img_url, headers=headers, timeout=10)
        with open(save_path, "wb") as f:
            f.write(response.content)
        return True
    except Exception as e:
        print(f"图片下载失败：{e}")
        return False
def crawl_target_imgs(driver):
    """批量爬取网页图片"""
    # 定位所有图片元素（根据实际网页修改选择器）
    img_elements = driver.find_elements(By.XPATH, '//div[@class="img-list"]//img')
    print(f"共找到{len(img_elements)}张图片")
    success_count = 0
    for index, img in enumerate(img_elements):
        # 获取图片真实地址
        img_url = img.get_attribute("src")
        if not img_url or img_url.startswith("data:image"):
            continue
        save_path = f"target_imgs/img_{index+1}.jpg"
        # 下载图片
        if download_img(img_url, save_path):
            success_count += 1
            print(f"第{success_count}张图片下载完成")
        time.sleep(random.uniform(0.5, 1))  # 随机延时，避免请求过快
    print(f"图片爬取完成，成功下载{success_count}/{len(img_elements)}张")
# 执行图片爬取
crawl_target_imgs(driver)
# 关闭浏览器
driver.quit()