火力与指挥控制2025，Vol.50Issue(10)：81-87,7.DOI:10.3969/j.issn.1002-0640.2025.10.010
结合多模态筛选和联合协同表示的SAR图像目标识别方法
Target Recognition in SAR Images via Combination of Selection of Multiple Modes and Joint Collaborative Representation
摘要
Abstract
The effectiveness of feature and classifier is improved specially for synthetic aperture radar(SAR)target recognition.In the feature extraction stage,multi-level descriptions of SAR images are obtained based on bidimensional empirical mode decomposition(BEMD),i.e.,bidimensional intrinsic mode functions(BIMFs).With the original image as a reference,the non-linear correlation information entropy(NCIE)is employed to select the optimal subset of BIMFs.In the classification stage,the joint collaborative representation is adopted to encode and represent the selected subset of BIMFs,obtaining the linear representation coefficient vectors under the optimal reconstruction condition.Finally,the target label of the test samples is determined based on the encoding errors from different classes.The experimental conditions are set up based on the MSTAR dataset,and the effectiveness of the proposed method is demonstrated through comparative verifications.关键词
合成孔径雷达/目标识别/二维经验模态分解/非线性相关信息熵/联合协同表示Key words
synthetic aperture radar/target recognition/bidimensional empirical mode decomposition/non-linear correlation information entropy/joint collaborative representation分类
计算机与自动化引用本文复制引用
李瑞芳,许洋洋,杨若璞,郭光立..结合多模态筛选和联合协同表示的SAR图像目标识别方法[J].火力与指挥控制,2025,50(10):81-87,7.基金项目
河南省2024年科技攻关基金资助项目(242102220015) （242102220015）
