New algorithm for colour image segmentation using hybrid k-means clustering

The traditional k-means algorithm is a classical clustering method which is widely used in variant application such as image processing, computer vision, pattern recognition and machine learning. However, the k-means method converges to one of many local minima. It is known that, the final result depends on the initial starting points (means). Generally initial cluster centres are selected randomly, so the algorithm could not lead to the unique result. In this paper, we present a new algorithm which includes three methods to compute initial centres for k-means clustering. First one is called geometric method which depends on equal areas of distribution. The second is called block method which segments the image into uniform areas. The last method is called hybrid and it is a combination between first and second methods. The experimental results appeared quite satisfactory. © 2012 Inderscience Enterprises Ltd.