模糊聚类算法是一种解决图像分割的常见算法,Stelios在模糊C均值聚类算法的基础上提出了FLICM算法,极大地改进了图像分割的效果。基于此,经过分析证明FLICM算法存在的不收敛问题,在此基础上改进了FLICM算法,并提出了结合遗传算法来解决因目标函数复杂度高而无法给出“闭合”迭代公式的问题。从结果来看,该算法不仅克服了FLICM算法不收敛的问题,而且取得了更好的图像分割效果,使得图像细节得到更充分的保留。 Fuzzy clustering algorithm is a common algorithm for image segmentation. Based on the fuzzy C-means clustering algorithm, Stelios proposed FLICM algorithm, which greatly improves the effect of image segmentation. Based on this, we prove that the FLICM algorithm does not converge, and improve the FLICM algorithm based on the above analysis. Finally, we propose a genetic algorithm to solve the problem that the “closed” iterative formula can not be given due to the high complexity of the objective function . From the result, the algorithm not only overcomes the problem that the FLICM algorithm does not converge, but also achieves a better image segmentation effect and allows the image details to be preserved more fully.