I’m teaching myself .NET c# through books and self-assigned projects for fun. I thought it could be a good experience to try and create my own image click captcha control from scratch. The kind where you identify “the right image” from a few options and click the right one (the cat or something) to identify yourself as human.
As I was trying to think of all the ways a script might learn its way around whatever I create, I considered the possibility that it could simply learn the right answers from trial-error and saving the filenames of each image. Eventually it’d learn which filenames were the “right ones”
I can’t think of any way to actually hide an image filename from a browser or source code, but renaming them every go-through isn’t practical either. Is there some way I could “render” the images in some sort of custom MIME type (is that the right question? i’m new sorry) each time they’re requested instead of just throwing out IMG SRC’s?
This might just be impossible, but figured I’d try asking the experts. Thanks for your time!
What you do is provide a proxy for the images:
What you do at your end it send the MIME header, then stream out the file. This way there is no direct relation between the image that is served and a particular filename.