Classification of Images by Using Distance Functions Defined on Intuitionistic Fuzzy Sets

Abstract

AbstractIntuitionistic fuzzy sets are mathematical structures which could be, together with operations, relations and functions defined on them, applied in many different spheres. In this paper, the distance functions defined on intuitionistic fuzzy sets are used to solve the problem of classification of images. Our specific problem is that we need to classify tire tread images into selected classes. Each class is characterized by its pattern. In the first step, the pre-processing of the image into the numeric data is done. The numerical data are represented by the specific vector. Then the values of membership function, non-membership function and hesitance margin for each coordinate of vector are calculated. As the last step of algorithm, the values of distance function between image and class patterns are computed. Classification is performed with the use of special function Sim whose values are dependent on distance function.