Clustering Using Annealing Evolution: Application to Pixel Classification of Satellite Images

In this article an efficient clustering technique, that utilizes an effective integration of simulated annealing and evolutionary programming as the underlying search tool, is developed. During the evolution process, data points are redistributed among the clusters probabilistically so that points that are farther away from the cluster center have higher probabilities of migrating to other clusters than those which are closer to it. The superiority of the new clustering algorithm over the widely used K-means algorithm and those based on simulated annealing and conventional evolutionary programming is demonstrated for some real life data sets. Another real life application of the developed clustering technique in classifying the pixels of a satellite image of a part of the city of Mumbai is also provided.