8.2 極驗滑動驗證碼的識別 · python3爬蟲筆記

### 1.目標通過程序識別并通過極驗驗證碼的驗證，包括 * 分析識別思路 * 識別缺口位置 * 生成滑動拖動路徑 * 模擬實現滑塊拼合通過驗證等步驟 ### 2.安裝 [selenium安裝](/1kai-fa-huan-jing-pei-zhi/12-qing-qiu-ku-de-an-zhuang/122-seleniumde-an-zhuang.md) [chromeDriver](/1kai-fa-huan-jing-pei-zhi/12-qing-qiu-ku-de-an-zhuang/123-chromedriverde-an-zhuang.md) ### 3.了解極驗驗證碼極驗驗證碼官網:[http://www.geetest.com/](http://www.geetest.com/) 它是一個專注于提供驗證安全的系統，主要驗證方式是拖動滑動拼合圖像。若圖像完全拼合，則驗證成功，即表單成功提交，否者重新驗證 ![](/assets/8.2.1.png) ### 4.極驗驗證碼特點 * 三角防護之防模擬 * 三角防護之防偽造 * 三角防護之防爆力 * 點擊一下，驗證只需要0.4秒 * 全平臺兼容，適用各種交互情景 * 面向未來，懂科技，更懂人性 ### 5.識別思路極驗官方后臺:[https://account.geetest.com/register](https://account.geetest.com/register) ![](/assets/8.2.2.png) 此按鈕為智能驗證按鈕，即同一個會話，一段時間內第二次點擊會直接通過驗證，如果智能識別不通過，則會彈出滑動驗證窗口，需要拖動滑塊拼合圖形完成二步驗證 ![](/assets/8.-2.3.png) 驗證成功后，按鈕會變成如下的所示狀態 ![](/assets/8.2.4.png) 識別驗證需要完成三步: * 模擬點擊驗證按鈕 * 識別滑動缺口的位置 * 模擬拖動滑動第一步:可以直接使用selenium模擬點擊按鈕第二步:識別缺口的位置，需要用到圖像的相關處理方法，需要用到圖像的相關處理方法第三步:拖動圖片 ### 6.初始化目標網址:[https://account.geetest.com/login](https://account.geetest.com/login) 初始化配置 ``` from selenium.webdriver.support.wait import WebDriverWait import config from selenium import webdriver class CrackGreetest(): def __init__(self): self.url = "https://account.geetest.com/login" self.browser = webdriver.Chrome() self.wait = WebDriverWait(self.browser,20) self.email = config.EMAIL self.passwrod = config.PASSWORD ``` ### 7.模擬點擊模擬點擊初始的驗證按鈕 ``` def get_geetest_button(self): ''' 獲取初始驗證碼按鈕 :return: 按鈕對象 ''' button = self.wait.until(expected_conditions.element_to_be_clickable((By.CLASS_NAME,'geetest_radar_tip'))) return button ``` ### 8.識別缺口 ``` def get_position(self,pos): """ 獲取驗證碼位置 :return: 驗證碼位置元組 """ img = self.wait.until(expected_conditions.presence_of_element_located((By.CLASS_NAME,pos))) time.sleep(2) location = img.location size = img.size top,bottom,left,right = location['y'],location['y']+size['height'],location['x'],location['x']+size['width'] return (top,bottom,left,right) def get_unFull_image(self,name): """ 獲取未完整驗證碼圖片 :param name: :return: 圖片對象 """ top,bottom,left,right = self.get_position(pos="geetest_canvas_img") print('驗證碼位置',top,bottom,left,right) screenshot = self.get_screenshot() unfull_captcha = screenshot.crop((left,top,right,bottom)) unfull_captcha.save(name) return unfull_captcha def get_full_image(self,name): ''' 獲取完整驗證碼圖片 :param name: :return: ''' # 這里要執行JavaScript腳本才能拿到完整圖片的截圖 show_Full_img1 = "document.getElementsByClassName('geetest_canvas_fullbg')[0].style.display='block'" self.browser.execute_script(show_Full_img1) show_Full_img2 = "document.getElementsByClassName('geetest_canvas_fullbg')[0].style.opacity=1" self.browser.execute_script(show_Full_img2) # 等待完整圖片加載 time.sleep(2) top, bottom, left, right = self.get_position(pos="geetest_canvas_fullbg") print('驗證碼位置', top, bottom, left, right) screenshot = self.get_screenshot() full_captcha = screenshot.crop((left, top, right, bottom)) full_captcha.save(name) return full_captcha ``` 比較兩張圖片 ``` def get_gap(self,image1,image2): """ 獲取缺口偏移量 :param image1: 不帶缺口的圖片 :param image2: 帶缺口的圖片 :return: 像素是否相同 """ # 缺口在滑塊右側，設定遍歷初始橫坐標left為59 left = 60 # 像素對比閾值 threshold = 60 for i in range(left, image2.size[0]): for j in range(image2.size[1]): rgb1 = image1.load()[i, j] rgb2 = image2.load()[i, j] res1 = abs(rgb2[0] - rgb1[0]) res2 = abs(rgb2[1] - rgb1[1]) res3 = abs(rgb2[2] - rgb1[2]) if not (res1 < threshold and res2 < threshold and res3 < threshold): return i - 7 # 返回缺口偏移距離，這里需測試幾次 ``` ### 9.模擬拖動 ``` def get_track(self,distanc): ''' x=v0*t+0.5*a*t*t v=v0+a*t 根據偏移量獲取移動軌跡 :param distanc: 偏移量 :return: 移動軌跡 ''' # 移動軌跡 track = [] # 當前位移 current = 0 # 減速閾值 mid = distanc * 4 / 5 # 計算間隔 t = random.randint(2,3)/10 # 初速度 v = 0 while current < distanc: if current < mid: # 加速度為正2 a = 2 else: # 加速度為負3 a = -3 v0 = v # 當前速度v = v0+at v=v0+a*t # 移動距離x=v0*t+1/2*a*t*t move = v0*t+1/2*a*t*t # 當前位移 current+=move # 加入軌跡 track.append(round(move)) return track def move_to_grap(self,slider,track): ''' 拖動滑塊到缺口處 :param slider: 滑塊 :param track: 軌跡 :return: ''' # 調用ActionChains的click_and_hold()方法按住拖動底部滑塊，遍歷運動軌跡獲取每小段位置距離，調用move_by_offset()方法移動此位移，最后調用release()方法松開鼠標 ActionChains(self.browser).click_and_hold(slider).perform() for x in track: ActionChains(self.browser).move_by_offset(xoffset=x,yoffset=0).perform() time.sleep(0.3) ActionChains(self.browser).release().perform() ``` ### 10.完整代碼 ``` import random from io import BytesIO from selenium.webdriver import ActionChains from selenium.webdriver.common.by import By from selenium.webdriver.support import expected_conditions from selenium.webdriver.support.wait import WebDriverWait import config,time from selenium import webdriver from PIL import Image class CrackGreetest(): def __init__(self): self.url = "https://account.geetest.com/login" self.browser = webdriver.Chrome() self.wait = WebDriverWait(self.browser,20) self.email = config.EMAIL self.password = config.PASSWORD def open(self): """ 打開網頁輸入用戶名密碼 :return: None """ self.browser.get(self.url) email = self.wait.until(expected_conditions.presence_of_element_located((By.ID,'email'))) password = self.wait.until(expected_conditions.presence_of_element_located((By.ID,'password'))) email.send_keys(self.email) password.send_keys(self.password) def get_geetest_button(self): ''' 獲取初始驗證碼按鈕 :return: 按鈕對象 ''' button = self.wait.until(expected_conditions.element_to_be_clickable((By.CLASS_NAME,'geetest_radar_tip'))) return button def get_screenshot(self): """ 獲取網頁截圖 :return: 截圖對象 """ screenshot = self.browser.get_screenshot_as_png() screenshot = Image.open(BytesIO(screenshot)) return screenshot def get_position(self,pos): """ 獲取驗證碼位置 :return: 驗證碼位置元組 """ img = self.wait.until(expected_conditions.presence_of_element_located((By.CLASS_NAME,pos))) time.sleep(2) location = img.location size = img.size top,bottom,left,right = location['y'],location['y']+size['height'],location['x'],location['x']+size['width'] return (top,bottom,left,right) def get_unFull_image(self,name): """ 獲取未完整驗證碼圖片 :param name: :return: 圖片對象 """ top,bottom,left,right = self.get_position(pos="geetest_canvas_img") print('驗證碼位置',top,bottom,left,right) screenshot = self.get_screenshot() unfull_captcha = screenshot.crop((left,top,right,bottom)) unfull_captcha.save(name) return unfull_captcha def get_full_image(self,name): ''' 獲取完整驗證碼圖片 :param name: :return: ''' # 這里要執行JavaScript腳本才能拿到完整圖片的截圖 show_Full_img1 = "document.getElementsByClassName('geetest_canvas_fullbg')[0].style.display='block'" self.browser.execute_script(show_Full_img1) show_Full_img2 = "document.getElementsByClassName('geetest_canvas_fullbg')[0].style.opacity=1" self.browser.execute_script(show_Full_img2) # 等待完整圖片加載 time.sleep(2) top, bottom, left, right = self.get_position(pos="geetest_canvas_fullbg") print('驗證碼位置', top, bottom, left, right) screenshot = self.get_screenshot() full_captcha = screenshot.crop((left, top, right, bottom)) full_captcha.save(name) return full_captcha def get_slider(self): """ 獲取滑塊 :return: 滑塊對象 """ slider = self.wait.until(expected_conditions.element_to_be_clickable((By.CLASS_NAME,'geetest_slider_button'))) return slider # # 判斷兩張圖片同一位置的像素是否相同。比較兩張圖的RGB的絕對值是否均小于定義的閾值threshold # # 如果絕對值均在閾值值內，則代表像素點相同，繼續遍歷，否者代表不相同的像素點，即缺口的位置 # def is_pixel_equal(self,image1,image2,x,y): # ''' # 判斷兩個像素是否相同 # :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 # ''' # # 取兩個圖片的像素點 # pixel1 = image1.load()[x,y] # pixel2 = image2.load()[x,y] # # 像素對比閾值 # 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 get_gap(self,image1,image2): """ 獲取缺口偏移量 :param image1: 不帶缺口的圖片 :param image2: 帶缺口的圖片 :return: 像素是否相同 """ # 缺口在滑塊右側，設定遍歷初始橫坐標left為59 left = 60 # 像素對比閾值 threshold = 60 for i in range(left, image2.size[0]): for j in range(image2.size[1]): rgb1 = image1.load()[i, j] rgb2 = image2.load()[i, j] res1 = abs(rgb2[0] - rgb1[0]) res2 = abs(rgb2[1] - rgb1[1]) res3 = abs(rgb2[2] - rgb1[2]) if not (res1 < threshold and res2 < threshold and res3 < threshold): return i - 7 # 返回缺口偏移距離，這里需測試幾次 def get_track(self,distanc): ''' x=v0*t+0.5*a*t*t v=v0+a*t 根據偏移量獲取移動軌跡 :param distanc: 偏移量 :return: 移動軌跡 ''' # 移動軌跡 track = [] # 當前位移 current = 0 # 減速閾值 mid = distanc * 4 / 5 # 計算間隔 t = random.randint(2,3)/10 # 初速度 v = 0 while current < distanc: if current < mid: # 加速度為正2 a = 2 else: # 加速度為負3 a = -3 v0 = v # 當前速度v = v0+at v=v0+a*t # 移動距離x=v0*t+1/2*a*t*t move = v0*t+1/2*a*t*t # 當前位移 current+=move # 加入軌跡 track.append(round(move)) return track def move_to_grap(self,slider,track): ''' 拖動滑塊到缺口處 :param slider: 滑塊 :param track: 軌跡 :return: ''' # 調用ActionChains的click_and_hold()方法按住拖動底部滑塊，遍歷運動軌跡獲取每小段位置距離，調用move_by_offset()方法移動此位移，最后調用release()方法松開鼠標 ActionChains(self.browser).click_and_hold(slider).perform() for x in track: ActionChains(self.browser).move_by_offset(xoffset=x,yoffset=0).perform() time.sleep(0.3) ActionChains(self.browser).release().perform() def login(self): """ 點擊登陸 :return: """ submit = self.wait.until(expected_conditions.element_to_be_clickable((By.CLASS_NAME, 'login-btn'))) submit.click() time.sleep(10) print('登錄成功') def crack(self): try: # 輸入用戶名 self.open() # 模擬點擊按鈕 button = self.get_geetest_button() button.click() # 獲取驗證碼圖片 image1 = self.get_unFull_image('unfull_captcha.png') # 獲取帶缺口的驗證碼圖片 image2 = self.get_full_image('full_captcha.png') # 對比兩張圖片像素點，獲取缺口位置，得到偏移距離 # 獲取缺口位置 distance = self.get_gap(image1,image2) print("缺口位置",distance) # 獲取移動軌跡 track = self.get_track(distance) print("滑動軌跡",track) # 模擬人的行為，拖動滑塊，完成驗證 slider = self.get_slider() # slider.click() # 拖動滑塊 self.move_to_grap(slider,track) success = self.wait.until( expected_conditions.text_to_be_present_in_element((By.CLASS_NAME, 'geetest_success_radar_tip_content'), '驗證成功')) print(success) self.login() except: self.crack() if __name__ == "__main__": crack = CrackGreetest() crack.crack() ```