CSFCM: An improved fuzzy C-Means image segmentation algorithm using a cooperative approach

• Published in: Expert systems with applications Vol. 166; p. 114063
• Main Authors: Abdellahoum, Hamza, Mokhtari, Nassim, Brahimi, Abderrahmane, Boukra, Abdelmadjid
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 15.03.2021; Elsevier BV
• Subjects: Algorithms; Biogeography based algorithm; Classification; Cooperation; Firefly algorithm; Fuzzy c-means; Genetic algorithm; Genetic algorithms; Heuristic methods; Histograms; Image classification; Image segmentation; Multiagent systems; Neural networks; Segmentation; Biogeography based algorithm; Segmentation; Genetic algorithm; Firefly algorithm; Classification; Fuzzy c-means
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2020.114063

Fuzzy c-means (FCM) is one of the most widely used classification algorithms specially in image segmentation. Like any algorithm, FCM has some drawbacks such as the choice of the number of clusters and the cluster’s center initialization. In this work, we propose new approaches to deal with these two drawbacks. We propose for the first problem two approaches. The first proposed approach exploits neural networks and the Xie and Beni index, while the second one exploits the histogram. Concerning the second problem, we propose a new metaheuristics cooperation approach using the Genetic Algorithm (GA), Biogeography Based Algorithm(BBO), and Firefly Algorithm (FA). This cooperation is managed by a multi-agent system allowing to determine automatically the fittest metaheuristics parameters. Finally, we propose to use a histogram-based version of FCM to reduce the execution time of the algorithm. Experimental results show that our proposed approach improves the performance of the basic FCM algorithm and outperforms other methods proposed in the literature. •Determination of the number of clusters in image segmentation.•Initialization of the clusters center (using a cooperative approach).•Ensuring good segmentation.