Color Image Segmentation Method Based on Region Growing and Ant Colony Clustering

Abstract

In this paper, a color image segmentation method is proposed based on region growing and ant colony clustering. By using the intensity similarity among pixels in the HSI color space and the edge information, seeds are first automatically selected. Then seeded region begins to grow based on color and spatial information. After that, ant colony clustering method is used to merge different regions of homogeneity. Finally good results of segmentation can be achieved. Experimental results show that this method is an effective method for image segmentation, which eliminates large quantities of cycles and calculations caused by traditional ant colony algorithm.