Edge detection on fuzzy near sets

Abstract

Fuzzy image processing involves various fuzzy approaches which help in understanding and processing of images as fuzzy sets. In the process of ED outlines, boundaries of an object and background in a image are detected. Hence edge detection is very important in digital image processing applications to obtain information from the frames as a forerunner step to feature extraction and object segmentation. In this paper the pixel values of an image is converted to fuzzy membership value matrix in which the columns are as fuzzy near sets and the corresponding membership and gray images are found using MATLAB. Using gray membership values some basic types of edge detection techniques are used and the corresponding fuzzy near membership matrices are constructed. The Euclidean and Hamming distances between two edge detection fuzzy value matrices are calculated and similarity measure is found. The higher value of similarity measure indicates that the images using two of the edge detection techniques are similar.