火力与指挥控制2025，Vol.50Issue(10)：94-100,7.DOI:10.3969/j.issn.1002-0640.2025.10.012
基于互补特征自适应熵加权决策融合的SAR目标识别方法
SAR Target Recognition Method via Adaptive Fusion of Complementary Features Based on Information Entropy
摘要
Abstract
Synthetic aperture radar(SAR)imaging is an important means of modern battlefield observation.For SAR image target recognition,this paper proposes a self-adaptive entropy weighted decision fusion method for complementary features.Contour descriptors,non-negative matrix factorization(NMF)and azimuth sensitivity are used to describe the multi-faceted features of SAR images.The three types of features have good complementarities and can better reflect various characteristics of the target.Based on joint sparse representation,the three types of features are jointly represented and their corresponding decision results are output.On this basis,the weights of the three features are calculated based on information entropy,and their decisions are fused accordingly to determine the target label of the test sample.The proposed method is validated based on MSTAR dataset,and the results show its effectiveness and robustness.关键词
合成孔径雷达/目标识别/互补特征/决策融合/联合稀疏表示/信息熵Key words
synthetic aperture radar/target recognition/complementary features/decision fusion/joint sparse representation/information entropy分类
电子信息工程引用本文复制引用
尹广举,李昆,王佳敏,赵鹏..基于互补特征自适应熵加权决策融合的SAR目标识别方法[J].火力与指挥控制,2025,50(10):94-100,7.
