爬虫进阶：验证码突破--7、滑动验证码破解
测试网址是网易cc，只是研究爬虫自动化
1、Iframe窗口切换
import time, re
from selenium import webdriver
chrome_option = webdriver.ChromeOptions()
driver = webdriver.Chrome(chrome_options=chrome_option)
driver.set_window_size(1440, 900)
driver.get("http://cc.163.com/")
driver.implicitly_wait(10)
if driver.find_element_by_id("browser-zoom-tips"):
driver.find_element_by_xpath("//div[@class='zoom-closeBt']").click()
#这里有个iframe，找到iframe父节点
ele=driver.find_element_by_xpath("//*[@id='js-login-urs']/iframe")
driver.switch_to_frame(ele)
driver.find_element_by_class_name("j-inputtext dlemail j-nameforslide").send_keys("********")
driver.find_element_by_class_name("j-inputtext dlpwd").send_keys("*********")
#driver.find_element_by_xpath("//div[@class='yidun_slider']").click()
2、滑动验证图片获取
如下图所示，在元素查找中能获得滑块和背景图的class属性，这样我们可以通过定位这个属性，将这两张图片下载下来
target = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'yidun_bg-img')))
template = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'yidun_jigsaw')))
target_link = target.get_attribute('src')
template_link = template.get_attribute('src')
target_img = Image.open(BytesIO(requests.get(target_link).content))
template_img = Image.open(BytesIO(requests.get(template_link).content))
target_img.save('target.jpg')
template_img.save('template.png')
local_img = Image.open('target.jpg')
3、获取匹配距离
def match(self, target, template):
img_rgb = cv2.imread(target)
img_gray = cv2.cvtColor(img_rgb, cv2.COLOR_BGR2GRAY)
template = cv2.imread(template, 0)
run = 1
w, h = template.shape[::-1]
print(w, h)
res = cv2.matchTemplate(img_gray, template, cv2.TM_CCOEFF_NORMED)
run = 1
# 使用二分法查找阈值的精确值
L = 0
R = 1
while run < 20:
run += 1
threshold = (R + L) / 2
print(threshold)
if threshold < 0:
print('Error')
return None
loc = np.where(res >= threshold)
print(len(loc[1]))
if len(loc[1]) > 1:
L += (R - L) / 2
elif len(loc[1]) == 1:
print('目标区域起点x坐标为：%d' % loc[1][0])
break
elif len(loc[1]) < 1:
R -= (R - L) / 2
return loc[1][0]
4、由于人滑动滑块是非匀速滑动的，有一定的加速，或者说是后移，所以在模拟滑动时，也需要加上这一块：
def get_tracks(self, distance):
"""
拿到移动轨迹，模仿人的滑动行为，先匀加速后均减速
匀变速运动基本公式：
①：v=v0+at
②：s=v0t+½at²
③：v²-v0²=2as
:param distance:需要移动的距离
:return:存放每0.3秒移动的距离
"""
distance += 20 # 先滑过一点，最后再反着滑动回来
# 初速度
v = 0
# 单位时间为0.3s来统计轨迹，轨迹即0.3s内的位移
t = 0.3
# 位移/轨迹列表，列表内的一个元素代表0.3s的位移
forward_tracks = []
# 当前位移
current = 0
# 到达mid值开始减速
mid = distance * 4 / 5
while current < distance:
if current < mid:
# 加速度越小，单位时间的位移越小，模拟的轨迹就越多越详细
a = 2
else:
a = -3
# 初速度
v0 = v
# 0.3秒时间内的位移
s = v0 * t + 0.5 * a * (t ** 2)
# 当前的位置
current += s
# 添加到轨迹列表,round()为保留一位小数且该小数要进行四舍五入
forward_tracks.append(round(s))
# 速度已经达到v，该速度作为下次的初速度
v = v0 + a * t
# 反着滑动到准确位置
back_tracks = [-3, -3, -2, -2, -2, -2, -2, -1, -1, -1] # 总共等于-20
return {'forward_tracks': forward_tracks, 'back_tracks': back_tracks}
5、以下就是完整的代码
from PIL import Image
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
import logging
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
import cv2
import numpy as np
from io import BytesIO
import time, requests
LocalDay = time.strftime("%Y-%m-%d")
logFile = r'D:\log\{0}.log'.format(LocalDay)
logging.basicConfig(level=logging.DEBUG,
format='%(asctime)s %(filename)s[line:%(lineno)d] %(levelname)s %(message)s',
datefmt='%a, %d %b %Y %H:%M:%S',
filename=logFile,
filemode='a+')
console = logging.StreamHandler()
console.setLevel(logging.INFO)
formatter = logging.Formatter('%(name)-12s: %(levelname)-8s %(message)s')
console.setFormatter(formatter)
logging.getLogger('').addHandler(console)
class CrackSlider():
"""
通过浏览器截图，识别验证码中缺口位置，获取需要滑动距离，并模仿人类行为破解滑动验证码
"""
def __init__(self):
self.url = 'http://cc.163.com/'
self.driver = webdriver.Chrome(chrome_options=webdriver.ChromeOptions())
self.wait = WebDriverWait(self.driver, 10)
self.zoom = 1
def open(self):
self.driver.maximize_window()
self.driver.get("http://cc.163.com/")
self.driver.implicitly_wait(10)
if self.driver.find_element_by_id("browser-zoom-tips"):
self.driver.find_element_by_xpath("//div[@class='zoom-closeBt']").click()
self.driver.find_element_by_xpath("//span[contains(.,'登录')]").click()
ele = self.driver.find_element_by_xpath("//div[@id='js-login-urs']/iframe")
self.driver.switch_to_frame(ele)
self.driver.find_element_by_name("email").send_keys("[email protected]")
self.driver.find_element_by_name("password").send_keys("huoyingchong64")
#self.driver.find_element_by_xpath("//div[@class='yidun_slider']").click()
def get_pic(self):
time.sleep(2)
target = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'yidun_bg-img')))
template = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'yidun_jigsaw')))
target_link = target.get_attribute('src')
template_link = template.get_attribute('src')
target_img = Image.open(BytesIO(requests.get(target_link).content))
template_img = Image.open(BytesIO(requests.get(template_link).content))
target_img.save('target.jpg')
template_img.save('template.png')
local_img = Image.open('target.jpg')
size_loc = local_img.size
self.zoom = 320 / int(size_loc[0])
def get_tracks(self, distance):
"""
拿到移动轨迹，模仿人的滑动行为，先匀加速后均减速
匀变速运动基本公式：
①：v=v0+at
②：s=v0t+½at²
③：v²-v0²=2as
:param distance:需要移动的距离
:return:存放每0.3秒移动的距离
"""
distance += 20 # 先滑过一点，最后再反着滑动回来
# 初速度
v = 0
# 单位时间为0.3s来统计轨迹，轨迹即0.3s内的位移
t = 0.3
# 位移/轨迹列表，列表内的一个元素代表0.3s的位移
forward_tracks = []
# 当前位移
current = 0
# 到达mid值开始减速
mid = distance * 4 / 5
while current < distance:
if current < mid:
# 加速度越小，单位时间的位移越小，模拟的轨迹就越多越详细
a = 2
else:
a = -3
# 初速度
v0 = v
# 0.3秒时间内的位移
s = v0 * t + 0.5 * a * (t ** 2)
# 当前的位置
current += s
# 添加到轨迹列表,round()为保留一位小数且该小数要进行四舍五入
forward_tracks.append(round(s))
# 速度已经达到v，该速度作为下次的初速度
v = v0 + a * t
# 反着滑动到准确位置
back_tracks = [-3, -3, -2, -2, -2, -2, -2, -1, -1, -1] # 总共等于-20
return {'forward_tracks': forward_tracks, 'back_tracks': back_tracks}
def match(self, target, template):
img_rgb = cv2.imread(target)
img_gray = cv2.cvtColor(img_rgb, cv2.COLOR_BGR2GRAY)
template = cv2.imread(template, 0)
run = 1
w, h = template.shape[::-1]
print(w, h)
res = cv2.matchTemplate(img_gray, template, cv2.TM_CCOEFF_NORMED)
run = 1
# 使用二分法查找阈值的精确值
L = 0
R = 1
while run < 20:
run += 1
threshold = (R + L) / 2
print(threshold)
if threshold < 0:
print('Error')
return None
loc = np.where(res >= threshold)
print(len(loc[1]))
if len(loc[1]) > 1:
L += (R - L) / 2
elif len(loc[1]) == 1:
print('目标区域起点x坐标为：%d' % loc[1][0])
break
elif len(loc[1]) < 1:
R -= (R - L) / 2
return loc[1][0]
def crack_slider(self):
slider = self.wait.until(EC.element_to_be_clickable((By.CLASS_NAME, 'yidun_slider')))
ActionChains(self.driver).click_and_hold(slider).perform()
for track in tracks['forward_tracks']:
ActionChains(self.driver).move_by_offset(xoffset=track, yoffset=0).perform()
time.sleep(0.5)
for back_tracks in tracks['back_tracks']:
ActionChains(self.driver).move_by_offset(xoffset=back_tracks, yoffset=0).perform()
ActionChains(self.driver).move_by_offset(xoffset=-4, yoffset=0).perform()
ActionChains(self.driver).move_by_offset(xoffset=4, yoffset=0).perform()
time.sleep(0.5)
ActionChains(self.driver).release().perform()
self.driver.find_element_by_id("dologin").click()
if __name__ == '__main__':
cs = CrackSlider()
cs.open()
target = 'target.jpg'
template = 'template.png'
cs.get_pic()
distance = cs.match(target, template)
tracks = cs.get_tracks((distance +7) * 0.66) # 对位移的缩放计算，*0.66本来应该是*cs.zoom的，不知道什么原因，比例总是错误
cs.crack_slider()
参考地址：
http://www.cnblogs.com/llhtjwq/p/9210694.html
以上代码我的个人电脑测试没有问题，比例是按照我自己的电脑量身定做。。