| 基于支持向量机的雷达高分辨距离像识别 |
| |
| 引用本文： | 沈丽民,李军显.基于支持向量机的雷达高分辨距离像识别[J].弹箭与制导学报,2009,29(2). |
| |
| 作者姓名： | 沈丽民 李军显 |
| |
| 作者单位： | 1. 河南科技大学电子信息工程学院,河南洛阳,471003
2. 河南科技大学电子信息工程学院,河南洛阳,471003;中国空空导弹研究院,河南洛阳,471009 |
| |
| 基金项目： | 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。
|
| 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 |
Radar High Range Resolution Profile Identification Based on SVM |
| |
| SHEN Limin,LI Junxian.Radar High Range Resolution Profile Identification Based on SVM[J].Journal of Projectiles Rockets Missiles and Guidance,2009,29(2). |
| |
| Authors: | SHEN Limin LI Junxian |
| |
| Affiliation: | School of 1 Electronic Information Engineering;Henan University of Science and Technology;Henan Luoyang 471003;China;2 China Airborne Missile Academy;Henan Luoyang 471009;China |
| |
| Abstract: | 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 results prove that the SVM classifier showed better performance than match... |
| |
| Keywords: | SVM HRRP radar target recognition match correlation method |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
