UTILIZATION OF INFORMATION MEASURE AS A MEANS OF IMAGE THRESHOLDING

The entropy method for image thresholding suggested by Kapur et al. ha s been modified and a more pertinent information measure of the image is obtained. Essentially this consists of viewing the image as a compo situm of two fuzzy sets correponding of the two classes with membershi p coefficient associated with each gray level a function of its freque ncy of occurrence as well as its distance from the intermediate thresh old selected. An extension of this technique to consider the semantic content of the image is also discussed. The superiority of the suggest ed method over artificial histograms modelled by Gaussian distribution s is demonstrated. Experimental results on several images are also pre sented to support the validity of the concepts used. (C) 1994 Academic Press, Inc.