摘要
车辆检测和车型识别是智能交通系统(Intelligent transportation system,ITS)中的一个重要方面,而目标识别是低分辨率雷达领域的一个难点。该文提出一种用多普勒雷达进行车型识别的方法,把车辆建模成包含多个散射中心的目标体,散射中心与雷达的距离与频谱能量有关,因此同一目标的频谱变化反映了该目标长高等轮廓特征。然后将有效的频谱特征结合主成分分析(Principal component and analysis,PCA)和线性判别分析(Linear discriminant analysis,LDA)进行降维,再利用支持向量机(Support vector machine,SVM)等分类器实现分型。文章对不同识别算法交叉验证的实验结果进行比较,表明基于PCA-LDA-SVM的车型识别算法效果理想,有广泛的应用前景。
Vehicle detection and recognition is important to the development of intelligent transportation system(ITS), but target recognition is a challenging problem for low-resolution radar. Hence, a vehicle recognition approach using Doppler radar is proposed, and the spec- trum variation of one vehicle reflects its outline. Then, the dimension of effective spectrum fea- ture can be reduced by the methods of principal component analysis (PCA) and linear discrimi- nant analysis (LDA). Vehicles can be classified into three types by classifier algorithms such as support vector machine(SVM), K-nearest neighbor(KNN). Finally, experimental results of different algorithms are compared by cross validation, and it shows that the algorithm based on PCA-LDA-SVM can achieve an ideal result.
出处
《数据采集与处理》
CSCD
北大核心
2012年第1期111-116,共6页
Journal of Data Acquisition and Processing
基金
国家高新技术发展计划("八六三"计划)基金(2008AA11Z203)资助项目
关键词
雷达目标识别
多普勒雷达
主成分分析
线性判别分析
支持向量机
radar target recognition
Doppler radar
principal component analysis
linear discriminant analysis
support vector machine