基于极化高分辨距离像的SVM目标识别方法

针对支持向量机（SVM）对单一雷达目标高分辨距离像数据源识别率较低和鲁棒性较差的问题，结合雷达目标极化信息，生成了雷达目标的极化综合距离像，提出了一种基于极化高分辨距离像的SVM目标识别方法。应用该方法分别对四类实测和仿真目标的ISAR成像数据进行了处理和识别实验，所得结果明显优于单一极化状态下的正确识别率，且鲁棒性更高。

SVM target identification method based on polarization HRRP

Aiming at the problem of the lower identification rate and worse robustness of the support vector machine (SVM) towards the data source of a single radar target high resolution range profile and integrating radar targets’ polarization information, targets’ polarization integration HRRP is produced, on the basis of which one SVM target identification method is proposed. This method is used for the processing and identifying experiments on four classes of real test and for emulating targets’ ISAR imaging data, and the clearly better and robust result is obtained. The correctness and validity of the method are proved.