The situation I have in mind is a .pdf document obtained from a scan, or at least such that some readers view it as an image (as opposed to text) and render it badly (see the screenshot below)
My personal user experience is with Evince in Ubuntu 12.04, but I guess this is common problem.
Is there something I can do inside Evince, like turn some option, etc.? Could you point to a valid alternative with a better rendering for images .pdf?
Using Evince on another Linux distribution and machine I get an acceptable rendering of the same document. Might this image rendering process be related to how the pdf reader interacts with the X-stuff setup?