I work for a company that does online marketing.
One of the things we do is submit our clients to various web directories.
Part of what we do is pull-in a web page and attempt to determine which image is the captcha image (anti-spam text).
One type of image that is particularly resistant to this process is what I call a "captcha generator" where the image that is displayed to the user is based upon a hash.
For example, if I go to a directory submission page that is using such a process for captchas, I can see the image. But if I copy the path to the image and try and see the image that way, I get a message saying "Could not fetch image phrase!!" or some such:
But, if you go to the actual submission page, the captcha works:
This is not a problem. I mean, we are not trying to automatically defeat the captcha -- just capture an image which is representative of the captcha image to display to the end-user who is processing the page.
Since I cannot visit the image directly, since the only way to see it is on the submission page (after it has rendered) -- my question is -- is there a way to capture the captcha image on the submission page somehow...perhaps by catching the image as it comes in as an HTTP response, or by some form of screen scraping? But I need sample source code ... particularly for captcha images that use a hash like the above example.
Again - this is not so that we can defeat the captcha image via OCR. It is just so that we can present the image to the end-user for selection. It need not even be the image that loads with the page originally. It could be an image from a page refresh if need be.
This would be happening on an ASP .NET page.