COMPARISON AND EVALUATION OF CLUSTER BASED IMAGE SEGMENTATION TECHNIQUES

Abstract

Image segmentation is the classification of an image into different groups. Numerous algorithms using different approaches have been proposed for image segmentation. A major challenge in segmentation evaluation comes from the fundamental conflict between generality and objectivity. A review is done on different types of clustering methods used for image segmentation. Also a methodology is proposed to classify and quantify different clustering algorithms based on their consistency in different applications. There are different methods and one of the most popular methods is k-means clustering algorithm. K-means clustering algorithm is an unsupervised algorithm and it is used to segment the interest area from the background. Enhanced k-means clustering is used to improve accuracy and efficiency of k means clustering algorithm. The number of clusters is changed for fuzzy c-mean algorithm. Subtractive cluster is used to generate the initial centers and these centers are used in k-means algorithm for the segmentation of image. Genetic algorithm is used for centroids in the given value K clusters (GAKM). GAKM is good for complex problems it retains best features. An outcome revealed that the accuracy and performance of GAKM is better than simple K-means and other clustering algorithms.