Fuzzy modified cuckoo search for biomedical image segmentation

In this article, a new method is proposed for biomedical image segmentation. The proposed method for biomedical image segmentation will be known as fuzzy modified cuckoo search (FMCS). This method falls under the category of unsupervised classification (i.e., clustering). In this work, the concept of a well-known metaheuristic method called cuckoo search is extended, modified, and combined with the modified type 2 fuzzy C-means algorithm, and the name is given accordingly. FMCS method uses a modified cuckoo search to find the optimum cluster centers based on fuzzy membership. The proposed FMCS technique fuses the idea of type 2 fuzzy sets with the MCS strategy, and it is applied in biomedical images segmentation. The proposed approach assists with deciding the clusters without having any affectability on the choice of the underlying centers. The quantity of the control variable for the MCS technique is very sensible contrasted with numerous other metaheuristics approaches. The MCS strategy can come to the global optima even subsequent to stalling out in a neighborhood optimum. The proposed method is applied to different biomedical images and compared with several standard optimization methods like genetic algorithm, particle swarm optimization, cuckoo search, etc. The proposed method does not suffer from the choice of initial cluster centers because it exploits the random behavior of the cuckoo search to initialize the cluster centers. Moreover, FMCS outperforms some of the standard methods in terms of the rate of convergence and other segmentation parameters. The proposed approach blends the type 2 fuzzy system in the modified cuckoo search procedure for efficient biomedical image segmentation. The superiority of the proposed method is verified by both quantitative and qualitative measures.