摘要
为利用合成孔径雷达(synthetic aperture radar,SAR)目标不同特征数据间的相关性与互补性,提出一种基于多特征的Tikhonov正则化核函数协同表示(multi-feature kernel collaborative representation-based classification with tikhonov regularization,MFKCRT)算法。采用美国运动和静止目标获取与识别(moving and stationary target acquisition and recognition,MSTAR)计划公开发布的SAR图像数据库进行实验,实现核函数变换空间上的多特征融合协同表示识别。实验结果表明:该算法相较于基本的协同表示,具有更优的可靠性与鲁棒性。
In order to exploit the correlation and complementarity between different feature data of synthetic aperture radar(SAR) target, a Tikhonov regularization kernel function collaborative representation algorithm based on m ultiple features is proposed. The SAR image database released by the moving and stationary target acquisition and recognition(MSTAR) program of the United States is used in the experiment. The multi-feature fusion collaborative representation and recognition on the kernel function transformation space is realized. Experimental results show that the proposed algorithm is more reliable and robust than the basic collaborative representation.
作者
刘苗苗
蒋宇帆
邢钉凡
Liu Miaomiao;Jiang Yufan;Xing Dingfan(Chongqing Optoelectronics Research Institute,Chongqing 400060,China;Military Representative Bureau in Chongqing,Chongqing 400060,China)
出处
《兵工自动化》
2022年第4期38-43,共6页
Ordnance Industry Automation
关键词
合成孔径雷达
自动目标识别
多特征
核函数
协同表示
SAR
automatic target recognition
multi-feature
kernel function
cooperative representation