International Journal of applied mathematics and computer science

online read us now

Paper details

Number 1 - March 2011
Volume 21 - 2011

KHM clustering technique as a segmentation method for endoscopic colour images

Mariusz Frąckiewicz, Henryk Palus

Abstract
In this paper, the idea of applying the k-harmonic means (KHM) technique in biomedical colour image segmentation is presented. The k-means (KM) technique establishes a background for the comparison of clustering techniques. Two original initialization methods for both clustering techniques and two evaluation functions are described. The proposed method of colour image segmentation is completed by a postprocessing procedure. Experimental tests realized on real endoscopic colour images show the superiority of KHM over KM.

Keywords
biomedical colour image segmentation, k-harmonic means technique, k-means technique

DOI
10.2478/v10006-011-0015-0