In this paper we propose a clustering technique that extracts sub-clusters based on a simple measure of isotropic symmetry. These sub-clusters are then used as building blocks to form final clusters of any arbitrary shape including concave ones through merging iteratively. The proposed method is tested on multi-spectral satellite imagery and a good result is obtained. Major advantages of this method are its simplicity and being free from initial guess about the cluster centres, shapes and the number of clusters. However, this algorithm is more suitable for mul-tivariate images even with very high spectral resolution.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.