Spatial kernel K-harmonic means clustering for multi-spectral image segmentation

Q. Li, N. Mitianoudis, T. Stathaki
2007 IET Image Processing  
The problem of image segmentation using intensity clustering approaches has been addressed in the literature. Grouping pixels of similar intensity to form clusters in an image has been tackled using a number of methods, such as the K-Means (KM) algorithm. The K-Harmonic Means (KHM) was proposed to overcome the sensitivity of KM to centre initialisation. In this paper, we investigate the use of a Spatial Kernel-based KHM (SKKHM) algorithm on the problem of image segmentation. Instead of the
more » ... Instead of the original Euclidean intensity distance, a robust Kernel-based K-Harmonic Means metric is employed to reduce the effect of outliers and noise. Spatial image information is also incorporated in the proposed clustering scheme, derived from Markov Random Field (MRF) modelling. An extension of the proposed algorithm to multi-spectral imaging applications is also presented. Experimental results on both single-channel and multi-channel images demonstrate the robust performance of the proposed SKKHM algorithm.
doi:10.1049/iet-ipr:20050320 fatcat:3xif3jfnhfaexh2gvo7ccuwv2u