An Image Segmentation of Fuzzy C-Means Clustering Based on the Combination of Improved Ant Colony Algorithm and Genetic Algorithm

This paper proposes a method of dynamic fuzzy clustering analysis based on improved ant colony algorithm. This method makes use of the great ability of ant colony algorithm for disposing local convergence, which overcomes sensitivity to initialization of fuzzy clustering method (FCM) and fixes on the numbers of clustering as well as the centers of clustering dynamically. This paper improves the traditional combination of the genetic algorithm and the ant colony algorithm, integrates the genetic algorithm with the ant colony algorithm, uses genetic algorithm´s rapidity and the overall astringency raised the ant group algorithm convergence rate, simultaneously, the regeneration enhanced the cluster precision using ant colony algorithm´s parallelism. At last, the application of the algorithm proposed to image segmentation and comparative experiments show that the mix algorithm has great ability of detection the fuzzy edge and exiguous edge.