I-FCM: A Cooperative Approach to Improve Fuzzy C-Mean: Application in Image Segmentation
Image segmentation is an important part of image processing. Several works have been proposed to create new methods or improve existing ones. Segmentation methods consist in grouping image pixels into regions according to criteria. The most widely used method in this field is fuzzy c-means(FCM). FCM is the result of the k-means technic with the application of fuzzy logic theory. This method has some drawbacks that can negatively influence the final results of segmentation. In this paper, we are interested in solving the problem of cluster centers initialization through the use of metaheuristics. we propose a new cooperative system that exploits three metaheuristics (Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Dragonfly Algorithm (DA)). An experimental study was conducted to evaluate our proposal, in which we compare our cooperative system with three metaheuristics (GA, DA, and PSO) and FCM. The final result shows that our proposal offers the best segmentation quality.
Keywords - FCM, Genetic algorithm, particle swarm optimization, Dragonfly algorithm, segmentation.