The Efficiency of the Adapted AntClust Algorithm for Satellite Images Clustering

Main Article Content

Ahmed BENYAMINA
Hadria FIZAZI

Abstract

This paper presents a novel algorithm for images satellite clustering using an adapted algorithm based on self-organization and the collective intelligence of ant colonies. The research aims to partition satellite images automatically by discovering the number of thematic classes in multispectral satellite images. Ants normally move in an array in one dimension and can carry objects. The attachment or removal of an object depends on a lot of similarity between this object and the heap objects. The probability that an ant takes the object is greater than leaving the object isolated. When an ant carries an image pixel, the probability that he deposits it as the element density of the same type in the neighbourhood is great. The experimental results of the AntClust adapted algorithm on satellite images can extract the correct class number.

Downloads

Download data is not yet available.

Article Details

How to Cite
BENYAMINA, A., & FIZAZI, H. (2011). The Efficiency of the Adapted AntClust Algorithm for Satellite Images Clustering. Malaysian Journal of Computer Science, 24(4), 168–179. Retrieved from https://sare.um.edu.my/index.php/MJCS/article/view/6577
Section
Articles