溫馨提示×

您好,登錄后才能下訂單哦!

密碼登錄×
登錄注冊(cè)×
其他方式登錄
點(diǎn)擊 登錄注冊(cè) 即表示同意《億速云用戶服務(wù)條款》

如何使用Selenium破解新浪微博的四宮格驗(yàn)證碼

發(fā)布時(shí)間:2021-08-10 11:44:22 來(lái)源:億速云 閱讀:122 作者:小新 欄目:開(kāi)發(fā)技術(shù)

這篇文章給大家分享的是有關(guān)如何使用Selenium破解新浪微博的四宮格驗(yàn)證碼的內(nèi)容。小編覺(jué)得挺實(shí)用的,因此分享給大家做個(gè)參考,一起跟隨小編過(guò)來(lái)看看吧。

在我們爬蟲(chóng)的時(shí)候經(jīng)常會(huì)遇到驗(yàn)證碼,新浪微博的驗(yàn)證碼是四宮格形式。

可以采用模板驗(yàn)證碼的破解方式,也就是把所有驗(yàn)證碼的情況全部列出來(lái),然后拿驗(yàn)證碼的圖片和這所有情況中的圖片進(jìn)行對(duì)比,然后獲取驗(yàn)證碼,再通過(guò)selenium自動(dòng)拖拽點(diǎn)擊,進(jìn)行破解。

如何使用Selenium破解新浪微博的四宮格驗(yàn)證碼

我們將驗(yàn)證碼四個(gè)點(diǎn)標(biāo)注為1234,那么所有的情況就是以下24種情況。

數(shù)字代表箭頭指向:

1234213431244321
1243214331424312
1342231432144123
1324234132414132
1423241334124213
1432243134214231

所有的情況就是以上24種。我們將這24中驗(yàn)證碼的情況放在一個(gè)文件夾內(nèi),當(dāng)我們?cè)诘卿浀臅r(shí)候用獲取的驗(yàn)證碼截圖去和所有的情況一一對(duì)比,然后獲取完全相同的驗(yàn)證碼,進(jìn)行點(diǎn)擊即可。代碼如下:

from selenium import webdriver
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.by import By
from selenium.common.exceptions import TimeoutException
from selenium.webdriver.common.action_chains import ActionChains
import time
from PIL import Image
from io import BytesIO
from os import listdir
USERNAME = ''
PASSWORD = ''
class CrackWeiboSlide():
  def __init__(self):
    self.url = 'https://passport.weibo.cn/signin/login'
    self.browser = webdriver.Chrome()
    self.wait = WebDriverWait(self.browser,20)
    self.username = USERNAME
    self.password = PASSWORD
  def __del__(self):
    self.browser.close()
  def open(self):
    """
    打開(kāi)網(wǎng)頁(yè)輸入用戶名密碼登錄
    :return: None
    """
    self.browser.get(self.url)
    username = self.wait.until(EC.presence_of_element_located((By.ID,'loginName')))
    password = self.wait.until(EC.presence_of_element_located((By.ID,'loginPassword')))
    submit = self.wait.until(EC.element_to_be_clickable((By.ID, 'loginAction')))
    username.send_keys(self.username)
    password.send_keys(self.password)
    submit.click()
  def get_position(self):
    """
    獲取驗(yàn)證碼的位置
    :return: 位置
    """
    try:
      img = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME,'patt-shadow')))
    except TimeoutException:
      print('未出現(xiàn)驗(yàn)證碼')
      self.open()
    time.sleep(2)
    location = img.location
    size = img.size
    top=location['y']
    bottom = location['y']+size['height']
    left = location['x']
    right = location['x']+size['width']
    return (top,bottom,left,right)
  def get_screenshot(self):
    """
    獲取截圖
    :return:截圖
    """
    screentshot = self.browser.get_screenshot_as_png()
    # BytesIO將網(wǎng)頁(yè)截圖轉(zhuǎn)換成二進(jìn)制
    screentshot = Image.open(BytesIO(screentshot))
    return screentshot
  def get_image(self,name):
    """獲取驗(yàn)證碼圖片"""
    top,bottom,left,right = self.get_position()
    print('驗(yàn)證碼位置',top,bottom,left,right)
    screenshot = self.get_screenshot()
    # crop()將圖片裁剪出來(lái),后面需要一個(gè)參數(shù)
    captcha = screenshot.crop((left,top,right,bottom))
    captcha.save(name)
    return captcha
  def detect_image(self,image):
    """
    匹配圖片
    :param self:
    :param image: 圖片
    :return: 拖動(dòng)順序
    """
    # 圖片所在的文件夾
    for template_name in listdir('templates/'):
      print('正在匹配',template_name)
      template = Image.open('templates/'+template_name)
      # 匹配圖片
      if self.same_img(image,template):
        # 將匹配到的文件名轉(zhuǎn)換為列表
        numbers = [int(number)for number in list(template_name.split('.')[0])]
        print('拖動(dòng)順序',numbers)
        return numbers
  def is_pixel_equal(self,image1,image2,x,y):
    """
    判斷兩個(gè)像素的相似度
    :param image1: 圖片1
    :param image2: 圖片2
    :param x: 位置x
    :param y: 位置y
    :return: 像素是否相同
    """
     # 取像素點(diǎn)
    pixel1 = image1.load()[x,y]
    pixel2 = image2.load()[x,y]
    # 偏差量等于60
    threshold = 60
    if abs(pixel1[0]-pixel2[0]) < threshold and abs(pixel1[1]-pixel2[1])<threshold and abs(pixel1[2]-pixel2[2])<threshold:
      return True
    else:
      return False
  def same_img(self,image,template):
    """
    識(shí)別相似的驗(yàn)證碼
    :param image: 準(zhǔn)備識(shí)別的驗(yàn)證碼
    :param template: 模板
    :return:
    """
    # 相似度閾值
    threshold = 0.99
    count = 0
    # 匹配所有像素點(diǎn)
    for x in range(image.width):
      for y in range(image.height):
        # 判斷像素
        if self.is_pixel_equal(image,template,x,y):
          count+=1
    result = float(count)/(image.width*image.height)
    if result>threshold:
      print('成功匹配')
      return True
    return False
  def move(self,numbers):
    """
    根據(jù)順序拖動(dòng),此處接收的參數(shù)為前面的驗(yàn)證碼的順序列表
    :param numbers:
    :return:
    """
    # 獲取四宮格的四個(gè)點(diǎn)
    circles = self.browser.find_elements_by_css_selector('.patt-wrap .patt-circ')
    print('-----------------',circles)
    dx = dy =0
    for index in range(4):
      circle = circles[numbers[index]-1]
      if index == 0:
        # 點(diǎn)擊第一個(gè)點(diǎn)
        ActionChains(self.browser).move_to_element_with_offset(circle,circle.size['width']/2,circle.size['height']/2).click_and_hold().perform()
      else:
        # 慢慢移動(dòng)
        times = 30
        for i in range(times):
          ActionChains(self.browser).move_by_offset(dx/times,dy/times).perform()
          time.sleep(1/times)
      if index == 3:
        # 松開(kāi)鼠標(biāo)
        ActionChains(self.browser).release().perform()
      else:
        # 計(jì)算下次的偏移
        dx = circles[numbers[index+1]-1].location['x'] - circle.location['x']
        dy = circles[numbers[index+1]-1].location['y'] - circle.location['y']
  def crack(self):
    """
    破解入口
    :return:
    """
    self.open()
    # 獲取驗(yàn)證碼圖片
    image = self.get_image('captcha.png')
    numbers = self.detect_image(image)
    self.move(numbers)
    time.sleep(10)
    print('識(shí)別結(jié)束')
if __name__ == '__main__':
  crack = CrackWeiboSlide()
  crack.crack()

設(shè)置自己的賬號(hào)密碼即可實(shí)現(xiàn)。

有時(shí)候會(huì)匹配不上,圖片相似度閾值達(dá)不到0.99以上,這個(gè)時(shí)候可能是我們收集的驗(yàn)證碼圖片過(guò)時(shí)了,重新開(kāi)啟圖片收集程序,運(yùn)行收集一下即可。

收集圖片程序代碼如下:

from selenium import webdriver
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.by import By
from selenium.common.exceptions import TimeoutException
import time
from PIL import Image
from io import BytesIO
from os import listdir
USERNAME = '18239831004'
PASSWORD = 'qweqweqwe'
class CrackWeiboSlide():
  def __init__(self):
    self.url = 'https://passport.weibo.cn/signin/login'
    self.browser = webdriver.Chrome()
    self.wait = WebDriverWait(self.browser,20)
    self.username = USERNAME
    self.password = PASSWORD
  def __del__(self):
    self.browser.close()
  def open(self):
    """
    打開(kāi)網(wǎng)頁(yè)輸入用戶名密碼登錄
    :return: None
    """
    self.browser.get(self.url)
    username = self.wait.until(EC.presence_of_element_located((By.ID,'loginName')))
    password = self.wait.until(EC.presence_of_element_located((By.ID,'loginPassword')))
    submit = self.wait.until(EC.element_to_be_clickable((By.ID, 'loginAction')))
    username.send_keys(self.username)
    password.send_keys(self.password)
    submit.click()
  def get_position(self):
    """
    獲取驗(yàn)證碼的位置
    :return: 位置
    """
    try:
      img = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME,'patt-shadow')))
    except TimeoutException:
      print('未出現(xiàn)驗(yàn)證碼')
      self.open()
    time.sleep(2)
    location = img.location
    size = img.size
    top=location['y']
    bottom = location['y']+size['height']
    left = location['x']
    right = location['x']+size['width']
    return (top,bottom,left,right)
  def get_screenshot(self):
    """
    獲取截圖
    :return:截圖
    """
    screentshot = self.browser.get_screenshot_as_png()
    # BytesIO將網(wǎng)頁(yè)截圖轉(zhuǎn)換成二進(jìn)制
    screentshot = Image.open(BytesIO(screentshot))
    return screentshot
  def get_image(self,name):
    """獲取驗(yàn)證碼圖片"""
    top,bottom,left,right = self.get_position()
    print('驗(yàn)證碼位置',top,bottom,left,right)
    screenshot = self.get_screenshot()
    # crop()將圖片裁剪出來(lái),后面需要一個(gè)參數(shù)
    captcha = screenshot.crop((left,top,right,bottom))
    captcha.save(name)
    return captcha
  # 獲取所有的驗(yàn)證碼
  def main(self):
    count = 0
    while True:
      name = str(count)+'.png'
      self.open()
      self.get_image(name)
      count+=1
if __name__ == '__main__':
  crack = CrackWeiboSlide()
  crack.main()

感謝各位的閱讀!關(guān)于“如何使用Selenium破解新浪微博的四宮格驗(yàn)證碼”這篇文章就分享到這里了,希望以上內(nèi)容可以對(duì)大家有一定的幫助,讓大家可以學(xué)到更多知識(shí),如果覺(jué)得文章不錯(cuò),可以把它分享出去讓更多的人看到吧!

向AI問(wèn)一下細(xì)節(jié)

免責(zé)聲明:本站發(fā)布的內(nèi)容(圖片、視頻和文字)以原創(chuàng)、轉(zhuǎn)載和分享為主,文章觀點(diǎn)不代表本網(wǎng)站立場(chǎng),如果涉及侵權(quán)請(qǐng)聯(lián)系站長(zhǎng)郵箱:is@yisu.com進(jìn)行舉報(bào),并提供相關(guān)證據(jù),一經(jīng)查實(shí),將立刻刪除涉嫌侵權(quán)內(nèi)容。

AI