Graphic objects are very widely and liberally used in
documents. Research magazines and news articles are just few
document examples where one can find ample usage of graphic
objects to elucidate the concept. The usage of graphic objects
greatly enhances the presentation of the concept and also gives
a good visual appeal. It also adds a lot of color and diversity to
an otherwise dull document.
Graphic objects also in general also consume the maximum
amount of ink while printing. This is true of color rich images
and also of images with rick background and texture. Generally
it would suffice to have low resolution images to deliver the
same understanding. When finer details are required the
document could always be referred to back in the computing
system. An eco-friendly option could also avoid printing of the
images during the initial reading.
The goal of this paper is to present the idea of reduced ink
printing. We present two simple approaches to robustly and
efficiently detect the graphic objects in a given page of the
document. This could then be used to mask out the graphic
object. A second option after detection is to re-embed the
graphic object in low resolution format.