Shulou(Shulou.com)06/01 Report--

This article mainly introduces the relevant knowledge of "python+opencv+selenium automatic login mailbox and realize sliding verification function". The editor shows you the operation process through an actual case. The operation method is simple, fast and practical. I hope that this article "python+opencv+selenium automatic login mailbox and realize sliding verification function" can help you solve the problem.

When we do automatic login, we always encounter all kinds of strange CAPTCHA, and the slider CAPTCHA is the most common one. If our program automatically enters the account password, we still need to manually slide the CAPTCHA, can it still be called automation?

Well, first of all, let me tell you my 'problem-solving steps'.

1. Use selenium to open the mailbox home page.

two。 Navigate to the account password box and type the account password.

3. Get the verification picture and use opencv processing to return the distance the slider should drag.

4. Create a mouse event to simulate dragging the slider to complete the verification.

Issues that need to be addressed by   :

1. The positioning of the page element.

two。 The frame nesting of the text box and CAPTCHA.

3.opencv processing verifies picture gap image matching and returns distance.

How 4.webdriver locates its elements when using xpath in a web page.

5. The size of the original picture is scaled in proportion to the actual size on the web page (scaling of the distance).

   source code:

# from selenium.webdriver.common.keys import Keysfrom selenium.webdriver import ActionChainsfrom selenium import webdriverimport requestsimport timeimport cv2# download image def download_img (url,filename): r = requests.get (url) with open (filename + ".png", "wb") as f: # access response content through r.content for image type Write the response content to f.write (r.content) print (filename + "download complete") def get_image () in yanzheng.png: # Why are global variables defined here? Because driver is in the function, the function will be closed after it has been run, and # the corresponding web page will also be closed, which is why many people on the Internet ask why the web page is automatically closed. Global driver driver= webdriver.Chrome () # get the browser object driver.get ("https://mail.qq.com/") # load Baidu homepage # window maximization operation if the window is too small Driver.maximize_window () time.sleep (2) # sleep for two seconds driver.find_element_by_xpath ("/ html/body/div/div [2] / div/div [1] / div/div [1] / div [2]"). Click () time.sleep (1) driver.switch_to.frame ("login_frame") # driver. Find_element_by_xpath ("/ / * [@ id=" switcher_plogin "]") .click () # enter the account password input=driver.find_element_by_xpath ("/ / * [@ id=" u "]") # locate the QQ account box time.sleep (1) input.send_keys ("zhanghao") # search box enter content input=driver.find_element_by_xpath ("/ / * [@ id=" p "] ") # locate QQ password box input.send_keys (" your password ") # search box input content print (" account password input is complete. ") # pause here, otherwise sliding verification will not be displayed, it should be a means to detect automated tools. (anti-crawl) time.sleep (1) driver.find_element_by_xpath ("/ / * [@ id=" login_button "]) .click () # Note here we need to jump to the subbox of the CAPTCHA (another nested) time.sleep (1) driver.switch_to.frame (" tcaptcha_iframe ") # webdriver is different from the browser xpath and cannot directly locate the attribute of the tag. # you need to navigate to webelement first, and then get to attributes! Bk = driver.find_element_by_xpath ("/ / * [@ id=" slideBg "]") .get_attribute ("src") print (bk) # get the background and slider address and download it locally. Key = driver.find_element_by_xpath ("/ / * [@ id=" slideBlock "]") .get_attribute ("src") print (bk) download_img (bk,filename= "bk") download_img (key Filename= "key") # Lock slider slider = driver.find_element_by_xpath ("/ / * [@ id=" tcaptcha_drag_thumb "]") # get the slider distance dis = get_distance () print (dis) # slider section No problem. It's done. Newact = ActionChains (driver) newact.click_and_hold (slider). Perform () newact.move_by_offset (xoffset=dis-20,yoffset=0). Perform () time.sleep (0.5) newact.release (). Perform () # to get the distance the slider should move. Def get_distance (): path = "bk.png" img = cv2.imread (path) path = "key.png" img2 = cv2.imread (path) imgContour = img.copy () print ("img.shape:", img.shape) imgGray = cv2.cvtColor (img, cv2.COLOR_BGR2GRAY) imgBlur = cv2.GaussianBlur (imgGray, (3,3), 1) imgCanny = cv2.Canny (imgBlur, 400500) imgGray2 = cv2.cvtColor (img2 Cv2.COLOR_BGR2GRAY) imgBlur2 = cv2.GaussianBlur (imgGray2, (3,3), 1) imgCanny2 = cv2.Canny (imgBlur2, 400,500) cv2.imshow ("O", imgCanny) # matching jigsaw result = cv2.matchTemplate (imgCanny, imgCanny2, cv2.TM_CCOEFF_NORMED) # Normalized cv2.normalize (result, result, 0,1, cv2.NORM_MINMAX,-1) min_val, max_val, min_loc Max_loc = cv2.minMaxLoc (result) print ("min_loc:", min_loc) print ("max_loc:", max_loc) # match results circle cv2.rectangle (imgContour, max_loc, (max_loc [0] + 135, max_loc [1] + 135), (0,0,255), 2) # the original image is 680mm 390 and the browser resize is 280mm 161 We only use width here. So it needs to be scaled to the same scale. Res = min_loc [0] / (680 / 280) cv2.imshow ("Canny Image", imgContour) # you can't use 0 here, because the picture window will always be displayed, and the program will not be able to return the distance for the slider function. Cv2.waitKey (100) print ("should slide distance for success.") Return resif _ _ name__ = "_ _ main__": get_image ()

Below the    is the running result, and the background images of two different CAPTCHA codes can be correctly identified. The red box is the red box marked after the code identifies the gap.

This is the end of the content about "python+opencv+selenium automatically logs in to the mailbox and implements the sliding authentication function". Thank you for reading. If you want to know more about the industry, you can follow the industry information channel. The editor will update different knowledge points for you every day.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.