基于直觉模糊核聚类的弹道中段目标识别方法

针对现有的模糊核聚类算法性能的问题，汲取直觉模糊c-均值聚类（Intuitionistic Fuzzy c-means，IFCM）算法的动态聚类特性优势，引入高斯核函数，改良归一化条件，提出直觉模糊核c-均值聚类（Intuitionistic Fuzzy Kernel c‐means，IFKCM）算法，并通过实际数据测试，证实了该算法的可行性和有效性。最后，根据弹道中段目标识别仿真系统的要求及弹道目标识别的特点，设计并实现了基于直觉模糊核c-均值聚类的弹道中段目标识别(Intuitionistic Fuzzy Kernel c-means-Target Recognition in Ballistic Midcourse，IFKCM-TRBM)原型系统，仿真实验及对比分析充分表明该原型系统的稳健可行性，为弹道中段目标识别提出了一种新的参考和尝试。

Technique for target recognition in ballistic midcourse based on intuitionistic fuzzy kernel clustering

A kernel based intuitionistic fuzzy clustering algorithm named IFKCM is proposed on the basis of analyzing the deficiency of the existing clustering algorithm. The new algorithm introduces Gauss kernel, improves the constraint conditions and uses the property superiority of dynamic clustering performance, which is also the superiority in Intuitionistic Fuzzy c-means (IFCM) algorithm. Then the experimental result has proved its effectiveness. Subsequently, according to the requirement of target recognition in ballistic midcourse simulation system and the character of the ballistic target recognition, the simulation system named Intuitionistic Fuzzy Kernel c-means —Target Recognition in Ballistic Midcourse (IFKCM-TRBM) is designed and realized. The results of simulation show that the system is reliable and can support the research of target recognition in ballistic midcourse.