Abstract: Fuzzy C-Means clustering algorithm（FCM） is one of the widely applied fuzzy algorithms at present, but FCM algorithm has some limitations. It is sensitive to initial clustering center and membership matrix and likely converges into the local minimum, so it can not get the best clustering results. Combining the Particle Swarm Optimization（PSO） and FCM, a new FCM clustering algorithm based on chaos particle swarm optimization（CPSO） is proposed. The algorithm can make use of the global optimization of PSO to jump out of local minimum, and can get a better clustering effect. In order to avoid stagnation of particles in the iteration, the algorithm introduces the chaotic variables, generates a chaos sequence based on the current global best position, and replaces randomly a particle of the particle swarm with the particle that has optimal-adaptive value in the chaos sequence. The algorithm is applied in image segmentation. The experimental results show that the new algorithm can segment the image effectively and properly, and has the good robustness to noises and good adaptability.

Key words: clustering, Fuzzy C-Means（FCM）, Particle Swarm Optimization（PSO）, Choas Particle Swarm Optimization（CPSO）, image segmentation

摘要： 模糊C均值聚类算法是目前使用最广泛的模糊聚类算法，但是该算法也有其局限性，比如在迭代过程中对初始值非常敏感，极容易陷入局部极小值，以至于得不到最佳聚类结果。将粒子群优化算法应用到模糊C均值聚类算法中，提出一种基于混沌粒子群的模糊C均值聚类算法。它能够利用粒子群算法强大的全局寻优能力避免算法收敛于局部极值，最大程度上达到全局最佳聚类结果。为了避免粒子在迭代过程中停滞，该算法引入了混沌变量，以当前的全局最优位置来产生一个混沌序列，用混沌序列中拥有最优适应值的粒子随机代替当前粒子群中的一个粒子。将基于混沌粒子群的模糊C均值聚类算法应用于图像分割中，实验结果表明该算法能够有效地分割图像，并具有良好的鲁棒性和适应性。

关键词: 聚类, 模糊C均值, 粒子群优化算法, 混沌粒子群优化算法, 图像分割