Grid captcha solution

#1
1629980118447.png

This method allows to solve any captcha where image can be divided into equal parts like reCAPTCHA V2 or hCaptcha. A grid is applied above the image. And you receive the numbers clicked boxes.

To solve the captcha you have to:
  1. Prepare the image and the instruction what to click. The instruction can be sent as text or as image. Don't forget to indicate the instruction language!
  2. Optionally: determine the grid and define it in your request with recaptchacols and recaptcharows parameters.
    If not defined we'll check the size of the image. If it's 300x300px we put 3x3 grid on it. If the size is different we put 4x4 grid.
  3. Submit a HTTP POST request to our API URL: http://2captcha.com/in.php including recaptcha parameter set to 1.
    Server accepts images in multipart or base64 format.
    Also you can provide instructions as an image using imginstructions parameter. For more info please check the full list of parameters in the table below.
  4. Server will return captcha ID or an error code if something went wrong.
  5. Make a 5 seconds timeout and submit a HTTP GET request to our API URL: http://2captcha.com/res.php to get the result.
    If captcha is already solved server will return the answer with numbers of grid cells that you need to click like: OK|click:3/8/9.
    Grid cells are numbered starting from number 1 from left to right and from top to bottom.
    If captcha is not solved yet server will return CAPCHA_NOT_READY result. Repeat your request in 5 seconds.
    If something went wrong server will return an error code.
  6. You simulate clicks on cells from the answer.
  7. If reCAPTCHA updates the set of images you can send us new request with an additional parameter previousID and the ID of previous request as a value. Then our worker will check only new images that were not selected by previous worker.
  8. Also you can proivde additional parameter can_no_answer in case if there's no images to click left. Worker will see a button to confirm that there's no corresponding images and API will return No_matching_images as answer.
 
Last edited: