Well you need to somehow emulate the equivalent of a web browser File->Save As. and then stat the contents of the downloaded files. Sounds possible but if you only want rendered images then you need to write or call a browser to simulate rendering the page. Maybe you could do this by looking at the browser cache or calling some external page caching software?
Probably also might want to look at the webserver access log as that will give you all the file hits needed for each page view. Easier than writing your own browser to parse css :)