摘要
提出了一种基于最优变换和聚类中心的雷达目标成像识别方法。该方法利用一种最优变换减少同类之间差异以及通过在子像空间选定一组最优聚类中心来增大异类之间差异 ,提高雷达目标识别率。仿真实验结果表明了该方法的有效性。
A novel approach of radar target recognition is proposed in this paper. The optimal transformation matrix is formed to reduce the difference between the same classes. The cluster centers with optimal separation between each other are selected in subprofile space to enhance the difference between different classes. The simulated results show the efficiency of proposed approach.
出处
《电波科学学报》
EI
CSCD
2002年第3期233-236,共4页
Chinese Journal of Radio Science
关键词
雷达
目标识别
一维距离像
最优聚类中心
最优变换
radar target recognition, range profile, optimal cluster centers, optimal transformation