Adaptive FCM Clustering Algorithm Based On Twin Multiple Population Genetic Evolution

Fuzzy C-Means clustering algorithm (FCM) is an important method to analyze MRI brain maps, but the determination of the initial clustering center will directly affect the effect of clustering. This problem is often solved by Multiple Population Genetic Algorithm (MPGA) Still, MPGA lacks global search capability and adaptiveness, so it is prone to premature convergence, and the optimized clustering center is not optimal. For this reason, this paper shows Adaptive FCM clustering algorithm based on twin multiple population genetic evolution (TMPG-AFCM). TMPG-AFCM uses the first proposed twin operator, which considers the problem of insufficient inter-population search ability and performs twin operation on initialized populations to improve the algorithm's search ability. At the same time, it is used to combine the fuzzy control concept to dynamically adjust the genetic probability to improve the algorithm's self-adaptability. Then it is used to optimize the initial clustering center of the FCM algorithm to improve the clustering effect of the algorithm on MRI brain maps. The analysis of the simulation results shows that this algorithm outperforms other alternative FCM algorithms in MRI brain map segmentation.