A New Cooperative Algorithm Based on Artificial Fish Swarm Algorithm and K-means for Image Segmentation

Image Segmentation is one of the most important techniques in graphic and image processing. Most of image segmentation methods are based on clustering algorithms. Dataclustering is an unsupervised classification technique and belongs to NP-hard problems. One of the methods for solvingNP-hard problems is applying swarm intelligence algorithms. Artificial fish swarm algorithm (AFSA) is one of the swarm intelligence algorithms which is working based on populationand random search. In this paper, a new cooperative algorithm based on AFSA and k-means is proposed for performing imagesegmentation based on multi-level thresholding. The proposed algorithm utilizes both global search ability of AFSA and localsearch ability of k-means. The proposed algorithm along with some other known algorithms has been applied for segmenting famous images and their efficiency has been compared with each other. Experimental results comparison shows acceptable efficiency of the proposed algorithm.

Keywords:

image segmentation , data clustering , artificial fish swarm algorithm , k-means , multilevel thresholding