Where are browser images cached ?

[problems@gmail]

I want to be able to capture images to eg. build my own
abbreviated versions of various sets of web-pages, edited
to a minumum.

ksnaphot [from KDE] can capture the whole screen OK.
But obviously the images are individually already available,
stored in the cache some where, after the page is saved
locally - bcause it can be reloaded off-line.

The raw-source shows obviouly the original URL only.
So there must be a mapping from 'that' to the ID allocated
by the caching mechanism ?

Thanks for any input,

== Chris Glur.