一种多类原型模糊聚类的初始化方法

模糊聚类是非监督模式分类的一个重要分支，在模式识别和图像处理中已经得到了广泛的应用，但现有模糊聚类算法大都需要聚类数的先验知识，而且对初始化极为敏感，从而限制了它们的实际应用。此外对于多类原型样本集的聚类分析，还需要事先已知原型的类型及相应数目。为了克服这些限制，本文提出一种聚类原型先验知识的获取方法，并用来初始化多类原型模糊聚类，取得了效好的效果。

An Initialization Method for Multi-Type Prototype Fuzzy Clustering

Fuzzy clustering is an important branch of unsupervised classification, and has been widely used in patternrecognition and image processing. However, most of exiting fuzzy clustering algorithms are sensitive to initialization, andstrongly depend on the number of clusters, which limits their applications. Moreover , it also needs to know the type and number of prototypes in advance in multi-type prototype fuzzy clustering. To overcome these limitation, a metthod for acquiring apriori knowledge about clustering prototype is proposed in this paper,which obtains better performance in initializing multitype prototype fuzzy clustering.