一种基于属性散射中心的SAR遮挡目标识别方法

针对合成孔径雷达（Synthetic Aperture Radar，SAR）图像遮挡目标识别问题，提出基于属性散射中心局部最优匹配的方法。综合考虑匹配对数量以及匹配对之间的差异，采用粒子群算法构建测试样本和模板样本的属性散射中心集之间的局部最优匹配关系。在此基础上，引入SAR遮挡目标带来的散射中心变化特性，定义稳健的相似度准则。基于散射中心匹配结果计算测试样本与各类模板的相似度进而判定其目标类别。在MSTAR数据集上的实验结果表明，所提方法在标准操作条件下对10类目标取得98.72%的平均识别率，在随机缺失、方向缺失两种遮挡模型下也取得优于对比方法的性能，证明了所提方法对于SAR遮挡目标识别的有效性。

A SAR occluded target recognition method based on attributed scattering centers

A locally optimal matching scheme of attributed scattering centers is proposed for recognition of occluded targets in synthetic aperture radar(SAR) images.Considering the number of matching pairs and differences between different matching pairs,the particle swarm optimization is adopted to establish the locally optimal matching between the attributed scattering center sets from the test sample and template sample.Accordingly,the properties of SAR occluded targets are used to define a robust similarity measure.The similarities between the test sample and different template samples are calculated to determine the target label.Experiment results on MSTAR dataset show that the proposed method can achieve an average recognition rate of 98.72% for 10 classes of targets under the standard operating condition,and its performance under random occlusion and directional occlusion is superior over the compared methods,which proves the validity of the proposed method for recognition of occluded targets in SAR images.