An evaluation of particle swarm optimization techniques in segmentation of biomedical images

Abstract

Image segmentation is a common image processing step to many computer vision applications with the purpose to segment pixels into different classes. As improved variants of particle swarm optimization (PSO) algorithms, the fractional-order Darwinian particle swarm optimization (FODPSO) and Darwinian particle swarm optimization (DPSO) have been proposed for image segmentation. The purpose of this paper is to compare the segmentation performance of PSO, DPSO, and FODPSO as parametric approaches to existing methods; namely the parametric fuzzy c-means (FCM) algorithm, and the non-parametric Otsu segmentation technique with application to five biomedical images. All PSO-based experiments are conducted with twenty runs to assess the effectiveness of PSO models. The universal quality index is used to evaluate the segmentation results. The obtained experimental results showed that particle swarm based algorithms outperformed both FCM and Otsu segmentation technique.