Military Representative Bureau in Chongqing
针对合成孔径雷达（Synthetic Aperture Radar, SAR）目标识别中方位角差距大的训练样本之间存在干扰的问题，对传统协同表示的表示字典进行改进，得到更适应于当前测试样本且能够降低弱相关原子对系统影响的自适应字典，基于此提出了一种自适应原子选择的核函数变换协同表示算法。在美国运动和静止目标获取与识别计划公开发布的SAR图像数据库上的实验结果表明，基于自适应原子选择的多特征核协同表示方法较基于全部训练样本字典的多特征核协同表示模型，降低了干扰原子的不良影响，提高了SAR目标识别的可靠性和鲁棒性。
Aiming at the disturbance of training samples with large aspect gap, a kernel function transformation collaborative representation algorithm based on adaptive atom selection is proposed, which is used for SAR target recognition. This method improves the representation dictionary in the traditional collaborative representation, and gets the adaptive dictionary which is more adaptable to the current test sample and can reduce the influence of the unrelated atom to the system. The experiments of SAR target recognition based on MSTAR datasets are carried out. The experimental results show that the multi-feature kernel collaborative representation based on adaptive atom selection is more effective than the multi-feature kernel collaborative representation model based on all training sample dictionary, which reduces the un-good influence of the interference atoms and further improves the recognition performance of the system.