EXTRACTING A DOCUMENT PAGE IMAGE FROM AN ELECTRONICALLY SCANNED IMAGE HAVING A NON-UNIFORM BACKGROUND CONTENT

摘要

A technique includes acquiring data representing a first image produced by electronically scanning a page against a background. The first image contains a non-uniform background content due at least in part to a variation introduced by the background being non-uniform. The technique includes extracting an image of the page from the first image, wherein the extraction includes characterizing the background content of the first image; identifying candidate pixels associated with the page based at least in part on the characterized background content; and based at least in part on the identified candidate pixels and a model for a boundary of the page, determining the boundary of the page.