Pepijn seems to have taken the words right out of my mouth (figuratively speaking, of course). If you don't know how to correctly identify the image source, all you need to do is access the page for the image itself, not the post on some image-sharing website displaying the image. For example:
Post displaying image
Image source
The URL for the post will return a little page, while the source will return the actual image.

I imagine that copying an image to your clipboard just stores the actual image data on your device temporary and it doesn't save any image links PMC can pull from. So for this to function PMC will need to completely switch how it handles most images (it just pulls from the image link you provide atm, while this requires PMC to actually save the image in its own database... I think). So this also means more bytes PMC needs to store.

All in all I would say it's a nice idea in theory but it got a few big problems to it, if I got all my information correct that is.