Thanks for replying, but I don't think that would work (correct me if I'm wrong). I tried building a scraper that downloads a webpage's HTML, follows all links

Thanks for replying, but I don't think that would work (correct me if I'm wrong). I tried building a scraper that downloads a webpage's HTML, follows all links, and keeps going until it finds at least 3 high-resolution images (using cURL to get their size), but of course it's not feasible. I can't spider a whole site looking for images, nor download all of them. Unfortunately I need to find some service that has already indexed the website.