基于Cloude分解的宽带双极化特征提取分析
|
摘要
针对双极化雷达体制下高速运动目标的分类识别,研究了宽带双极化散射矩阵的特征提取算法以及其在目标分类识别中的应用。根据3类弹道仿真目标在同极化和交叉极化状态下的高分辨一维像,利用Cloude分解提取目标的H/α/P特征,提取的特征表征散射中心的散射机理,利用支持向量机(SVM)算法进行分类识别,得到不同方位角和不同信噪比下的识别结果。仿真结果表明,高速运动的弹道目标可以利用其双极化特征进行分类识别,且不同方位角域范围内识别率不同,信噪比越大识别准确率越高。
Aiming at the classification and recognition of high-speed moving targets in the dual-polarized radar system,the feature extraction algorithm of broadband dual-polarized scattering matrix and its application in target classification and recognition are studied.According to the high resolution range profile(HRRP)on the same polarization and cross polarization state of three types of ballistic simulation targets.Cloude decomposition is used to extract the H/α/P features of the target,and the extracted features characterize the scattering mechanism of the scattering center.The SVM algorithm is used for classification and identification,and the recognition results under different azimuth angles and different SNRs are obtained.The simulation results show that the high-speed ballistic targets can be classified and identified by dual-polarization features,and recognition rates are different in different azimuth ranges.The recognition accuracy increases with the improve of SNR.
Aiming at the classification and recognition of high-speed moving targets in the dual-polarized radar system,the feature extraction algorithm of broadband dual-polarized scattering matrix and its application in target classification and recognition are studied.According to the high resolution range profile(HRRP)on the same polarization and cross polarization state of three types of ballistic simulation targets.Cloude decomposition is used to extract the H/α/P features of the target,and the extracted features characterize the scattering mechanism of the scattering center.The SVM algorithm is used for classification and identification,and the recognition results under different azimuth angles and different SNRs are obtained.The simulation results show that the high-speed ballistic targets can be classified and identified by dual-polarization features,and recognition rates are different in different azimuth ranges.The recognition accuracy increases with the improve of SNR.
引文
[1]常宇亮.瞬态极化雷达测量、检测与抗干技术研究[D].长沙:国防科技大学,2010.
[2]戴幻尧,刘勇,周波,等.极化雷达测量体制研究进展[J].现代雷达,2014(1):1-5.
[3]代大海,廖斌,肖顺平,等.雷达极化信息获取与处理的研究进展[J].雷达学报,2016,5(2):143-155.
[4]王雪松.雷达极化技术研究现状与展望[J].雷达学报,2016,5(2):119-131.
[5]俞鼎峰,喻荣梅,马超等.基于一维距离像的序列的导弹目标进动参数估计[J].电子测量技术,2015,38(2):55-59.
[6]CLOUDE S R,POTTIER E. An entropy based classificationschemeforlandapplicationsof polarimetric SAR[J].IEEE Transactions on GRS,1997,35(1):549-557.
[7]徐一凡,刘爱芳,徐辉,等.基于改进三分量模型的全极化SAR图像分类[J].电子测量技术,2017,40(12):220-227.
[8]肖顺平,王雪松,代大海,等.极化雷达成像处理及应用[M].北京:科学出版社,2013.
[9]郭雷.宽带目标极化特征提取与核方法识别研究[D].长沙:国防科学技术大学,2009.
[10] CLOUDE S R,POTTIER E.A review of target decomposition theorems in radar polarimetry[J].IEEE Transactions on GRS,1996,34(2):498-517.
[11]白晓静.基于Cloude分解的特征参数分析及快速替代方法[D].成都:电子科技大学,2010.
[12]邢远见,陈翼,祝欢,等.基于支撑向量机的地杂波虚警抑制技术[J].现代雷达,2016,38(8):34-38.
[13]张维坤,叶伟,李国靖,等.基于相关向量机的SAR图像飞机目标分类方法研究[J].电子测量技术,2017,40(1):151-154.
[14] FULLER D F. Phase history decomposition for efficient scatter classification in SAR imagery[D].Dayton:Air Force Institute of Technology,2011.
[15] DONG Y H,BRUCE C F,TICEHURST C.A new decomposition of polarization signatures[J].IEEE Transactions on GRS,1998,36(3):933-939.
[2]戴幻尧,刘勇,周波,等.极化雷达测量体制研究进展[J].现代雷达,2014(1):1-5.
[3]代大海,廖斌,肖顺平,等.雷达极化信息获取与处理的研究进展[J].雷达学报,2016,5(2):143-155.
[4]王雪松.雷达极化技术研究现状与展望[J].雷达学报,2016,5(2):119-131.
[5]俞鼎峰,喻荣梅,马超等.基于一维距离像的序列的导弹目标进动参数估计[J].电子测量技术,2015,38(2):55-59.
[6]CLOUDE S R,POTTIER E. An entropy based classificationschemeforlandapplicationsof polarimetric SAR[J].IEEE Transactions on GRS,1997,35(1):549-557.
[7]徐一凡,刘爱芳,徐辉,等.基于改进三分量模型的全极化SAR图像分类[J].电子测量技术,2017,40(12):220-227.
[8]肖顺平,王雪松,代大海,等.极化雷达成像处理及应用[M].北京:科学出版社,2013.
[9]郭雷.宽带目标极化特征提取与核方法识别研究[D].长沙:国防科学技术大学,2009.
[10] CLOUDE S R,POTTIER E.A review of target decomposition theorems in radar polarimetry[J].IEEE Transactions on GRS,1996,34(2):498-517.
[11]白晓静.基于Cloude分解的特征参数分析及快速替代方法[D].成都:电子科技大学,2010.
[12]邢远见,陈翼,祝欢,等.基于支撑向量机的地杂波虚警抑制技术[J].现代雷达,2016,38(8):34-38.
[13]张维坤,叶伟,李国靖,等.基于相关向量机的SAR图像飞机目标分类方法研究[J].电子测量技术,2017,40(1):151-154.
[14] FULLER D F. Phase history decomposition for efficient scatter classification in SAR imagery[D].Dayton:Air Force Institute of Technology,2011.
[15] DONG Y H,BRUCE C F,TICEHURST C.A new decomposition of polarization signatures[J].IEEE Transactions on GRS,1998,36(3):933-939.