一种基于多维特征的弹道目标识别方法  

Ballistic target recognition method based on multi-dimensional features

作　　者：张维 张福涛 张仕元[1] ZHANG Wei;ZHANG Futao;ZHANG Shiyuan(Nanjing Research Institute of Electronics Technology,Nanjing 210039,China)

机构地区：[1]南京电子技术研究所,南京210039

出　　处：《空天预警研究学报》2023年第6期391-397,共7页JOURNAL OF AIR & SPACE EARLY WARNING RESEARCH

摘　　要：针对传统基于目标RCS序列统计特征的识别方法对当前突防技术弹道目标识别有效性下降的问题,通过提取反映弹道目标运动机理和分形特性差异的多维度特征,提出了一种多维特征的弹道目标识别方法.仿真数据的特征可视化结果和识别置信度结果表明,本文提出的方法能较好地区分出真假弹头目标,显著提高复杂干扰场景下的识别准确率.The traditional recognition methods based on target RCS sequence statistical features of target have an attenuated effectiveness for ballistic target recognition in current penetration techniques.In order to solve the aforementioned problem,a multi-dimensional feature recognition method for ballistic targets is proposed by extracting the multi-dimensional features that reflect the differences in ballistic target motion mechanism and fractal characteristics.The feature visualization result and recognition confidence result of simulation data show that the method proposed in the paper can better distinguish the real and fake warhead targets,significantly improving the recognition accuracy in complex interference scenarios.

关 键 词：弹道目标识别 多维特征 运动机理 分形特性 

分 类 号：TN957[电子电信—信号与信息处理]