基于稀疏分解的海面微动目标识别  

Sea surface micro-moving target recognition based on sparse decomposition

作　　者：黄瀚仪 胡仕友[2] 郭胜龙 李珊君[1] 舒勤[1] HUANG Hanyi;HU Shiyou;GUO Shenglong;LI Shanjun;SHU Qin(College of Electrical Engineering,Sichuan University,Chengdu 610000,China;Beijing Huahang Radio Measurement Institute,Beijing 100013,China)

机构地区：[1]四川大学电气工程学院,四川成都610000 [2]北京华航无线电测量研究所,北京100013

出　　处：《系统工程与电子技术》2023年第4期1016-1023,共8页Systems Engineering and Electronics

摘　　要：海洋环境下杂波较强,慢速微弱目标的多普勒频率往往会落入海杂波多普勒频宽中,传统动目标检测方法难以检测出目标回波。为了解决此类问题,依据海杂波与目标在震荡属性和稀疏特性上的差异,首先利用可调Q小波变换算法分别获得对应的自适应完备字典。然后,运用形态成分分析算法得到对应的目标稀疏系数和杂波稀疏系数;再把稀疏系数与各自的自适应字典相乘得到目标分量与杂波分量。最后,在雷达对海探测数据集下验证了算法的有效性。Clutter in the marine environment is strong,and the Doppler frequency of slow and weak targets often falls into the sea clutter Doppler bandwidth.It is difficult for the classic moving target detection methods to detect target echoes.In order to address such problems,this paper uses the tunable Q-factor wavelet transform(TQWT)algorithm to obtain the corresponding self-adaptive complete dictionaries according to the difference between the sea clutter and the target in the oscillation properties and sparse characteristics,and then uses the morphological component analysis(MCA)algorithm to obtain the corresponding target sparse coefficients and clutter sparse coefficients.Then,the target components and clutter components are obtained by multiplying the sparse coefficients with their respective adaptive dictionaries.Finally,the effectiveness of the algorithm is verified by using the radar sea detection dataset.

关 键 词：海杂波 目标识别 稀疏分解 可调Q小波变换 

分 类 号：TN95[电子电信—信号与信息处理]