基于GSVD的核不相关辨别子空间与雷达目标识别
doi: 10.3724/SP.J.1146.2008.00384
Radar Target Recognition Based on Kernel Uncorrelated Discriminant Subspace of GSVD
-
摘要: 该文提出了一种基于广义奇异值分解的核不相关辨别子空间算法,并将其用于高分辨距离像雷达目标识别。新算法结合广义奇异值分解与核方法的优点,有效地解决了传统方法面临的矩阵奇异问题,同时进一步改善了目标的类可分性。其次,依据Fisher准则导出了距离像总散度矩阵零空间中不含有有用辨别信息的结论。利用这一结论,可以在求解核不相关最优辨别矢量之前对各散度矩阵进行预降维,以减小后续运算的计算复杂度。对3类飞机目标实测数据的识别结果表明了所提方法的有效性。Abstract: A Kernel Uncorrelated Discriminant Subspace (KUDS) method based on Generalized Singular Value Decomposition (GSVD) for radar target recognition is proposed. The new method combines with the advantage of GSVD and kernel trick, which can effectively overcome the limitation of traditional linear methods in solving singular problem, but also improve the class separability further. In addition, a conclusion from Fishers criterion that there exists no useful discriminative information in the null space of the range profile population scatter matrix is derived, which can be used to reduce the dimensionality of original scatter matrices as well as the computation complexity of the following operation of solving kernel optimal discriminant vectors. Experimental results based on three measured airplanes data confirm the effectiveness of the proposed method.
-
袁莉, 刘宏伟, 保铮. 雷达高分辨距离像分类器的参数自适应学习算法[J].电子与信息学报.2008, 30(1):198-202浏览[2]周代英, 杨万麟. 雷达目标一维距离像识别中的最优因式分析子空间法[J].电子与信息学报.2007, 29(10):2341-2345浏览[3]刘华林, 杨万麟. 基于QR 分解的广义辨别分析用于雷达目标识别[J]. 红外与毫米波学报, 2007, 26(3): 205-208.Liu H L and Yang W L. Radar target recognition based ongeneralized discriminant analysis of QR decompostition[J]. J.Infrared Millim. Waves, 2007, 26(3): 205-208.[4]Yu X L, Wang X G, and Liu B Y. A direct kerneluncorrelated discriminant analysis algorithm[J].IEEE SignalProcessing Letters.2007, 14(10):742-745[5]Jin Z, Yang J Y, and Hu Z S, et al.. Face recognition based onthe uncorrelated discriminant transformation[J]. PatternRecognition, 2001, 34(7): 1405-1416.[6]杨静宇, 金忠, 胡钟山. 具有统计不相关性的最佳鉴别特征空间的维数定理[J]. 计算机学报, 2003, 26(1): 110-115.Yang J Y, Jin Z, and Hu Z S. A theorem on dimensionality ofthe uncorrelated optimal discriminant features space[J].Chinear Journal of Computers, 2003, 26(1): 110-115.[7]Liang Z Z and Shi P F. Uncorrelated discriminant vectorsusing a kernel method[J].Pattern Recognition.2005, 38(1):307-310[8]Howland P and Park H. Generalizing discriminant analysisusing the generalized singular value decomposition[J].IEEETrans. on Pattern Analysis and Machine Intelligence.2004,26(8):995-1006[9]Baudat G and Anouar F. Generalized discriminant analysisusing a kernel approach[J].Neural Computation.2000, 12(10):2385-2404[10]Sch.kopf B, Smola A, and Mller K. Nonlinear componentanalysis as a kernel eigenvalue problem[J]. NeuralComputation, 1998, 10(5): 1299-1319.
点击查看大图
计量
- 文章访问数: 7896
- HTML全文浏览量: 110
- PDF下载量: 794
- 被引次数: 0