SAR Target Feature Extraction and Automatic Recognition Based on KFD Criterion

In this paper, a method of SAR ATR (Synthetic Aperture Radar Automatic Target Recognition) based on KFD (Kernel Fisher Discriminant) is described. It consideres the projections of samples onto an optimal vector derived by Fisher's discriminant in nonlinear space as sample features, then SVM classifier is used to implement target identification. Experimental results show better performance of identification and less running time. Moreover, we conclude that good experimental results can be obtained even if no target azimuth information is knawn. [