Automatic Determination Number of Cluster for Multi Kernel NMKFCM Algorithm on Image Segmentation

Abstract

In image analysis, image segmentation performed an essential role to get detail information about image. Image segmentation is suitable in many applications like medicinal, face recognition, pattern recognition, machine vision, computer vision, video surveillance, crop infection detection and geographical entity detection in map. FCM is famous method used in fuzzy clustering to improve result of image segmentation. FCM doesn’t work properly in noisy and nonlinear separable image, to overcome this drawback, Multi kernel function is used to convert nonlinear separable data into linear separable data and high dimensional data and then apply FCM on this data. NMKFCM method incorporates neighborhood pixel information into objective function and improves result of image segmentation. New proposed method used RBF kernel function into objective function. RBF function is used for similarity measure. New proposed algorithm is effective and efficient than other fuzzy clustering algorithms and it has better performance in noisy and noiseless images. In noisy image, find automatically required number of cluster with the help of Hill-climbing algorithm.