基于高分辨雷达距离像降维双谱特征的舰船目标识别OA

In order to address the issue of using high-resolution radar range profiles(HRRPs)for ship target recognition,this paper investigates the differences in the performance of ship target recognition using four reduced dimensionality integration bispectra features including axially integrated bispectra(AIB),radially integrated bispectra(RIB),circularly integrated bispectra(CIB)and selected bispectra.Firstly,different reduced dimensional-ity integration bispectra are used to extract the features from ship target HRRP,and then support vector machine(SVM)is used to classify and recognize the extracted features.Finally,the effects of different feature vector lengths and training sample sizes on the recognition performance of different reduced dimensionality integration bispectra are analyzed through simulation and experiment.Simulation and experimental results show that when achieving optimal recognition performance,different reduced dimensionality integration bispectra require differ-ent feature vector lengths and training sample sizes to be selected.