针对弹道目标的极化高分辨一维距离像,研究宽带极化特征提取算法,分析所提取特征表征的目标散射特性及其在目标分类识别中的应用。通过弹道目标全极化高分辨一维距离像,构建极化散射矩阵,估计弹道目标的极化相干矩阵,再对其进行Cloude 分解,提取H /α/ A / P特征,依此对目标各散射中心的散射机理进行判定,同时分析各特征对于弹道目标可分性的强弱,利用SVM算法进行分类识别,同时考虑了信噪比对识别结果的影响,用仿真实验验证了算法的有效性。

Wideband polarization feature extraction algorithm can be used to analyze the high resolution range profile (HRRP) of ballistic targets, The target scattering characteristics of the extracted features are analyzed and their applications in target classification and recognition are studied. Through the polarimetric HRRP of ballistic targets, construct polarization scattering matrices, estimate polarization coherence matrices, Cloude decomposition algorithm is used to extract features from the polarization coherence matrices, extracted H / α / A / P features are used for scattering mechanism identification of scattering centers. At the same time, the strength of each feature for the ballast target separability are analyzed, Using SVM algorithm for classification and recognition, Analyze the relationship between SNR and recognition rate. Finally, the simulation is used to verify the effectiveness of the algorithm.

TN957