基于支持向量机的雷达高分辨距离像识别  

Radar High Range Resolution Profile Identification Based on SVM

作　　者：沈丽民[1] 李军显[1,2] 

机构地区：[1]河南科技大学电子信息工程学院,河南洛阳471003 [2]中国空空导弹研究院,河南洛阳471009

出　　处：《弹箭与制导学报》2009年第2期231-234,共4页Journal of Projectiles,Rockets,Missiles and Guidance

基　　金：航空基金(20060151003);河南科技大学人才科学研究基金(09001121)资助

摘　　要：以支持向量机(SVM)作为分类器,研究了雷达目标高分辨距离像(HRRP)分类法,设计了相应的预处理算法,提出一种结合留一法和单一验证法的参数选择新方案。基于三种雷达目标的HRRP数据,比较了SVM分类法和匹配相关分类法性能,实验结果表明SVM算法在目标姿态的稳定性、对训练集大小的稳定性和抗噪能力方面都占有相当优势。Based on the support vector machine（SVM） was adopted in radar target high range resolution profile （HRRP） classification, a preprocessing algorithm was designed and a new SVM model selection scheme combining leave-one-out （LOO） method and one-validation method was presented. The classification performance of the SVM method was compared with the maximum correlation method on the classification of three types HRRP target, which experimental resuits.prove that the SVM classifier showed better performance than match correlation method （MCM）classifier in target-aspect and training set-size stability, as well in anti-noise ability.

关 键 词：支持向量机 高分辨距离像 雷达目标识别 匹配相关法 

分 类 号：TN957.52[电子电信—信号与信息处理]