In an larger LaTeX document there are often only some pages with color content (mainly figures) and the remaining ones are only black and white.
Because printing costs for color pages are much higher than for black and white it would be good to be able to extract all pages with color and print them separately. The first step for this is to be able to detect if a page contains color or not. This could be in a form of an text list of page number suitable to be read by a PDF page extraction script (using e.g.
A simple solution sufficient for many people would be to detect all pages which contain a
figure and assume that only these have color. However, a general solution would be nice. Only color elements which are printed should be taken into account, while e.g. the color frames around link by
hyperref should not. It is OK that the solution would disable these for the detection.