AUTOMATED IMAGE ENHANCEMENT ALGORITHM BASED ON OPTIMIZATION OF GENERALIZED FUZZY ENTROPIES (GFEs)

摘要

This paper presents a novel automated entropy-domain algorithm for contrast enhancement based on fuzzy sets theory. In view of performing contrast enhancement a novel class of generalized fuzzy entropies (GFEs) is introduced as an extension of the classical definition of entropy of a fuzzy set. The definition of entropy contrast is also carried out for evaluating the uniformity of distribution of the informational content over the total gray-level range of the image. In the first stage of the algorithm the image is transferred into the fuzzy domain and a set of optimal transformation parameters, based on Yager class of involutive membership functions, is calculated in accordance with maximum entropy principle. Using the notion of entropy contrast as an optimization criterion, the overall optimal gray-level modification parameter is obtained. The proposed algorithm provides a self-contained framework for dealing with image processing problems in an informational context.