First, the preparatory work
The use of Selenium, browser Chrome, and configure ChromDriver
Second, analysis
1. Click the Verify button analog : Selenium can be used directly to complete.
2. Identify the slider gap positions : the position of the first observation image and the peripheral edge of the notch, using a contrast detection to identify its original position notch.
At the same time the gap and get original pictures, set a contrast threshold, then traversed two pictures to find the same pixel position of the pixel RGB gap exceeds this threshold. That gap position
3. Analog drag the slider : extremely increases the mechanical test track identification and speed detection, only completely simulate human operations to be verified. Usually the first trajectory acceleration, and deceleration.
Third, code implementation
1. Initialization
Meizu login registration page using the test, https: //i.flyme.cn/register ?. Here initialize some configuration
import time from io import BytesIO from PIL import Image from selenium import webdriver from selenium.webdriver import ActionChains from selenium.webdriver.common.by import By from selenium.webdriver.support.ui import WebDriverWait from selenium.webdriver.support import expected_conditions as EC phone = '18888888888' class GeetestSpider(): def __init__(self): self.url = "https://i.flyme.cn/register?" self.browser =webdriver.Chrome(executable_path=r'D:\Google\Chrome\Application\chromedriver') self.wait = WebDriverWait(self.browser, 20) self.phone = phone
2. Click simulation
def get_button(self): """ Obtaining initial verification button, simulate a click : Return: button object """ button = self.wait.until(EC.element_to_be_clickable((By.CLASS_NAME, 'geetest_radar_tip'))) return button
3. Identify gaps
Two matching pictures, inconsistencies gap is the position before and after the acquisition. Use Selenium get the picture element to obtain the location and width and height, and then get the screenshot of the entire page, and then cut out of the picture can be.
def get_screen_image(self): """ Get screenshots : Return: Screenshot Object """ screen_img = self.browser.get_screenshot_as_png() screen_img = Image.open(BytesIO(screen_img)) return screen_img def get_position(self): """ Acquiring location verification codes : Return: the location codes ancestral """ img = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'geetest_canvas_img'))) 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_geetest_img(self, name='captcha.png'): """ Get the captcha : Return: Picture Object, Image Object """ top, bottom, left, right = self.get_position() print('验证码位置：', top, bottom, left, right) screen_img = self.get_screen_image() # 剪裁图片 captcha = screen_img.crop((left, top, right, bottom))
captcha.save(name) return captcha def get_slider(self): """ Get the slider : Return: Slider Object """ slider = self.wait.until(EC.element_to_be_clickable((By.CLASS_NAME, 'geetest_slider_button'))) return slider
After acquiring the slider slider, calling click () method to trigger the click, the gap like the picture presented. Then call get_geetest_img () method to get the second chapter picture.
The acquired two pictures are assigned to img1, img2. For picture contrast, each coordinate point traversed images, acquiring RGB image data corresponding to two pixel points. If the difference between the RGB data both within a certain range, the pixels representing the same two continue to compare the next point. If the difference exceeds a certain range, pixel represents not the same, that is, the gap position.
def equal_rgb(self, img1, img2, x, y): """ Analyzing two pixels are the same : Param img1: Picture 1 : Param img2: Picture 2 : Param x: x position : Param y: y position : Return: whether the same "" " # Get two pictures pixel PIXEL1 = img1.load () [the X-, the y-] pixel2 = img2.load()[x, y] 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, img1, img2): """ Gets offset : Param img1: picture without gaps : Param img2: picture notched : Return: Offset """ left = 60 for i in range(left, img1.size[0]): for j in range(img1.size[1]): if not self.equal_rgb(img1,img2,i,j): left = i return left return left
4. Analog drag the slider
To simulate human operation, drag the slider speed should not be uniform, and should not be kept at a certain speed shaking up and down. The rate of change of the slider should be at the beginning of the acceleration, deceleration will close the gap, we imagine to manually drag the slider.
Acceleration represented by a slider, is represented by the current velocity v, represented by the initial velocity v0, indicated by a moving distance x, t represents time movement, satisfies the following relationship:
x = v0 * t + a * t * t / 2
v = v0 + a * t
def get_track(self, d): """ The offset acquisition trajectory : Param d: offset amount : Return: trajectory, each movement "" " # Trajectory Track = [] # current displacement Current = 0 # decelerated starts decelerating 2/3 offset, the offset amount of the displacement is provided deviation = d * 2 / 3 # Interval t = 0.2 # Initial velocity V = 0 while current < d: if current < deviation: # 加速阶段 a = 2 the else : # deceleration phase a = -2 # Initial velocity V0 = V # current velocity V = V0 + A * T # displacement move = v0 * t + a * t * t / 2 # Current displacement Current + = Move # is added to the track, after the decimal track.append (round (Move)) return Track def move_slider(self, slider, track): """ Move the slider to follow the trajectory gap : Param slider: Slider : Param track: track :return: "" " # Mouse Hold slider ActionChains (self.browser) .click_and_hold (Slider) .perform () for I in Track: # traversed track element, each movement corresponding to a displacement ActionChains (self.browser) .move_by_offset (xoffset = I, yoffset = 0) .perform () the time.sleep ( 0.5 ) # After the move is complete, release the mouse ActionChains (self.browser) .release (). perform ()
Verification is successful:
The complete code
# _*_ coding=utf-8 _*_ import time from io import BytesIO from PIL import Image from selenium import webdriver from selenium.webdriver import ActionChains from selenium.webdriver.common.by import By from selenium.webdriver.support.ui import WebDriverWait from selenium.webdriver.support import expected_conditions as EC # 初始化 class GeetestSpider(): def __init__(self): self.url = "https://i.flyme.cn/register" self.browser = webdriver.Chrome(executable_path=r'D:\Google\Chrome\Application\chromedriver') self.wait = WebDriverWait(self.browser, 20) def get_button(self): """ 获取初始验证按钮，模拟点击 :return:按钮对象 """ button = self.wait.until(EC.element_to_be_clickable((By.CLASS_NAME, 'geetest_radar_tip_content'))) return button def get_screen_image(self): """ 获取网页截图 :return: 截图对象 """ screen_img = self.browser.get_screenshot_as_png() screen_img = Image.open(BytesIO(screen_img)) return screen_img def get_position(self): """ 获取验证码的位置 :return: 验证码位置元祖 """ img = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'geetest_canvas_img'))) 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_geetest_img(self, name='captcha.png'): """ 获取验证码图片 :return: 图片对象,Image对象 """ top, bottom, left, right = self.get_position() print('验证码位置：', top, bottom, left, right) screen_img = self.get_screen_image() # 剪裁图片 captcha = screen_img.crop((left, top, right, bottom)) captcha.save(name) return captcha def get_slider(self): """ 获取滑块 :return:滑块对象 """ slider = self.wait.until(EC.element_to_be_clickable((By.CLASS_NAME, 'geetest_slider_button'))) return slider def equal_rgb(self, img1, img2, x, y): """ 判断两个像素点是否相同 :param img1: 图片1 :param img2: 图片2 :param x: 位置x :param y: 位置y :return: 是否相同 """ # 获取两张图片的像素点 pixel1 = img1.load()[x, y] pixel2 = img2.load()[x, y] 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, img1, img2): """ 获取偏移量 :param img1:不带缺口的图片 :param img2: 带缺口的图片 :return: 偏移量 """ # 直接从滑块的右侧开始遍历 left = 60 for i in range(left, img1.size[0]): for j in range(img1.size[1]): if not self.equal_rgb(img1, img2, i, j): left = i return left return left def get_track(self, d): """ 根据偏移量获取运动轨迹 :param d: 偏移量 :return: 运动轨迹,每次移动距离 """ # 运动轨迹 track = [] # 当前位移 current = 0 # 开始减速的偏移量,设位移达到偏移量的2/3时开始减速 deviation = d * 2 / 3 # 间隔时间 t = 0.2 # 初始速度 v = 0 while current < d: if current < deviation: # 加速阶段 a = 2 else: # 减速阶段 a = -2 # 初始速度 v0 = v # 当前速度 v = v0 + a * t # 位移 move = v0 * t + a * t * t / 2 # 当前位移 current += move # 添加到轨迹，保留整数 track.append(round(move)) return track def move_slider(self, slider, track): """ 按照轨迹移动滑块至缺口 :param slider:滑块 :param track: 轨迹 :return: """ # 鼠标按住滑块 ActionChains(self.browser).click_and_hold(slider).perform() for i in track: # 遍历轨迹元素，每次移动对应位移 ActionChains(self.browser).move_by_offset(xoffset=i, yoffset=0).perform() time.sleep(0.5) # 移动完成后，松开鼠标 ActionChains(self.browser).release().perform() def crack(self): """ 模拟验证的各种操作 :return:None """ # 打开网站，输入注册手机号 self.browser.get(self.url) self.wait.until(EC.presence_of_element_located((By.ID, 'phone'))) # 点击验证 button = self.get_button() button.click() # 获取验证码图片 img1 = self.get_geetest_img('captcha1.png') # 获取滑块 slider = self.get_slider() slider.click() # 获取带缺口的图片 img2 = self.get_geetest_img('captcha2.png') # 获取缺口位置 gap = self.get_gap(img1, img2) print("缺口位置：", gap) # 减去缺口位移 gap -= 6 # 获取轨迹 track = self.get_track(gap) print("轨迹：", track) # 拖动滑块 self.move_slider(slider, track) success = self.wait.until( EC.text_to_be_present_in_element((By.ID, 'geetest_success_radar_tip_content'), '验证成功')) # 失败重试 if not success: self.crack() if __name__ == '__main__': crack = GeetestSpider() crack.crack()