Uncertainty in multilevel image thresholding using Atanassov’s intuitionistic fuzzy sets

The problem of segmentation in spite of all the work over the last decades is still an important research field. Moreover, during the past years, fuzzy logic theory has been successfully applied to image thresholding. Considering that for segmentation purposes, in most cases, image pixels have an inherent ambiguity in the predicate that they must fulfill to belong to an object, which results in the experts uncertainty in assigning the pixel to that object, Atanassovpsilas intuitionistic fuzzy sets (A-IFSs) are a relevant and interesting extension since, uncertainty is one of the underlying ideas behind this theory. In this paper we describe a thresholding technique using A-IFSs. This approach uses Atanassov´s intuitionistic index values for representing the uncertainty of the expert when assigning the pixel to the background or to the object. The general framework of this approach and its natural extension to multilevel thresholding are presented. Segmentation results and their performance evaluation are presented.