Performance Of SEMI Supervised Fuzzy Clustering Algorithm For Change Detection In Remotely Sensed Multitemporal Images

Abstract: Fuzzy c-means (FCM) clustering algorithm widely used in image segmentation. However, its computational efficiency and wide spread reputation, the FCM algorithm does not take the spatial information of pixels into consideration, and thus may possibly result in low robustness to noise and less accurate segmentation. In this paper, semi supervised fuzzy clustering (SEMI-FCM) algorithm is presented for fuzzy segmentation of magnetic resonance (MR) images. To estimate the intensity in homogeneity, the global intensity is introduced into the logical limited intensity clustering algorithm and takes the local and global intensity information into account. The proposed method has been successfully applied to recorded GMRF + ICM with desirable results. Our results show that the proposed algorithm can effectively reduce the false alarm of image. Comparisons with other HTNN and EM demonstrate the better performance of the proposed SEMI- FCM algorithm.

Keywords: Fuzzy c-means, spatial-contextual, semi- supervised FCM, k- means, GMRF + ICM.