Classification of image data using gradient-based fuzzy c-means with mercer kernel

Abstract

A novel approach for the classification of image signals for image retrieval using Gradient-Based Fuzzy C-Means with Mercer Kernel (GBFCM-MK) is proposed and presented in this paper. The proposed classifier is a FCM-based algorithm which utilizes the Mercer Kernel to exploit the statistical nature of the image data to improve the classification accuracy. Experiments and results on various data sets demonstrate that the proposed classification algorithm outperforms 21.7% - 24% in accuracy in comparison with conventional algorithms such as the traditional Fuzzy C-Means (FCM), Gradient-based Fuzzy C-Means (GBFCM), and GBFCM with Divergence Measure (GBFCM(DM). © 2008 Springer-Verlag Berlin Heidelberg.