Merging pixels' location and illumination levels information for getting automatic fuzzy perceptual image segmentation algorithms

摘要

Monochrome digital images have a discrete set of illumination levels. So, results of monochrome images segmentation algorithms could be notably improved if human being's ability, for distinguishing these levels, would be considered within this type of algorithms. Although illumination levels provide a lot of information about the objects appearing within an image, to be considered within segmentation algorithms, gray levels constituting each illumination level has to be specified. As a part of our ongoing work for getting fuzzy perceptual segmentation algorithms, here we present a system for recognizing the image's illumination levels wherein, with the aim of simplifying the problems and automating the process, besides the analysis of image's gray levels histogram, location of the pixels associated with each histogram peak is also considered.