基于PCA-LVQ的雷达目标一维距离像识别  

1-D range profile recognition of radar targets using PCA-LVQ

作　　者：张小英[1] 王宝发[1] 刘铁军[1] 

机构地区：[1]北京航空航天大学电子工程系,北京100083

出　　处：《系统工程与电子技术》2005年第8期1373-1375,共3页Systems Engineering and Electronics

摘　　要：研究了宽带高分辨雷达目标识别问题。基于目标一维距离像,提出主成分分析(principal componentanalysis,PCA)和学习向量量化(learning vector quantization,LVQ)相结合的识别方法。即采用主成分分析方法进行数据压缩,提取目标特征,然后利用学习向量量化人工神经网络,训练集进行训练,建立识别模板库。最后对三种飞机模型高分辨回波数据进行识别,结果表明,经过PCA-LVQ网络处理后,目标维数和网络规模均大大降低,且系统具有良好的识别性能。The problem of high resolution range radar target recognition is studied. Firstly the target features are extracted from the one-dimension range profiles of radar targets by principal component analysis （ PCA）, then the training sets are trained by artificial neural network （ANN） of the learning vector quantization （LVQ）, thus the templates are obtained, The experiments of recognition using the high resolution radar data of three kinds of aircraft models are performed. The results demonstrate that the method is feasible.

关 键 词：模式识别 主成分分析 学习向量量化 一维距离像 

分 类 号：TN957.52[电子电信—信号与信息处理]