The use of non-free images on wikipedia is bad enough as it is. Allowing users to maintain the copyright would make Wikipedia "just another website", not the collaborative resource it is now. If you really want to maintain copyright, you have to do on your own personal portfolio, where you are fully responsible for anything posted. Even Flickr, Deviantart, Picasa, and all the other photo sites take some copyrights for themselves. The fact that they hide it in a Submission Policy doesn't make you safe. But if you have a good photo and want to share it, not caring where it pops up, then Wikipedia is for you. Some commercial practices for Wikipedia have developed over time, but your exclusive copyrights are gone forever when you submit something under one of its very permissive licenses. You can still offer your photos on a non-exclusive commercial license but that's it, you can't sell your exclusive copyright anymore, because you waived it for a common good.

I take it that you want to extract and compare features of the actual jpeg image, regardless of quality. There are many ways to do that and none of them includes filesize comparisons or the like. You could look in the JPEG Standard and try to filter out compression by just reading the base of every 8x8 block (that's the one that shouldn't be compressed) and compare these values for similarity. However you should aim for more advanced image recognition and comparison algorithms, for example the ones used on TinEye. Most of these algorithms come from the field of AI, but they're quite simple generally.