摘要
针对弹道单目标识别中,传感器资源利用率低、识别耗时长的缺点,采用群目标理论,提出了基于特征敏感度的弹道群目标聚类识别方法。该方法首先对目标群进行聚类,然后选取目标特征组合对各个分群进行敏感度计算,确立各分群的威胁度。同时,通过比较某个特定阶段各类特征的优劣性,为特征优化组合提供指导意见,优化了目标特征组合,提高了识别效率。最后,仿真实验验证了方法的可行性。
Aiming at low utilization rate and time-consuming identification of sensor resources of single ballistic target recognition,this paper puts forward the ballistic target group clustering recogni-tion method based on feature sensitivity .The keys to this method include clustering the target group and forming some small groups firstly,and then selecting the target feature combination to calculate the sensitivity for each small group,at the last establishing threat degree.At the same time,by com-paring the advantages and disadvantages of a particular stage,this paper brings out guiding opinions on feature optimization and combination,optimizes the feature combination and improves the efficien-cy of identification.In the end,the feasibility of the method is proven by simulation results.
出处
《装备学院学报》
2015年第3期52-57,共6页
Journal of Equipment Academy
基金
部委级资助项目
关键词
特征敏感度
群目标
聚类识别
feature sensitivity
group target
clustering recognition
