Image Segmentation Through Fuzzy Clustering: A Survey

Abstract

In modern years, image processing is a vast area for research. Image segmentation is the most popular part of image processing which divides the image into number of segments to analyze the better quality of image. It is used to detect objects and boundaries in images. Main goal of image segmentation is to change the representation of image into the more meaningful regions. Image segmentation results in a set of segments that covers the whole image or curves that are extracted from the image. In this paper, different image segmentation techniques and algorithms are presented, and clustering is one of the techniques that is used for segmentation. Fuzzy c-means clustering (FCM) algorithm is presented in this paper for image segmentation. On the basis of literature reviewed, several problems are analyzed in previously FCM, and the problems have been overcome by modifying the objective function of the previously FCM, and spatial information is incorporated in objective function of FCM. Fuzzy c-mean clustering is also known as soft clustering. The techniques that are explained in this survey are segmentation of the noisy medicinal images along spatial probability, histogram-based FCM, improved version of fuzzy c-means (IFCM), fuzzy possibilistic c-means (FPCM), possibilistic c-means (PCM), and possibilistic fuzzy c-means (PFCM) algorithms are to be explained in further sections on the basis of literature review. Moreover, several recent works on fuzzy c-means using clustering till 2017 are presented in this survey.