基于Mean Shift算法的多目标识别与仿真  

Multi-target Recognition and Simulation Based on Mean Shift Algorithm

作　　者：韩光威 顾力伟 刘咏 HAN Guang-wei;GU Li-wei;LIU Yong(The 723 Institute of CSIC,Yangzhou 225101,China)

机构地区：[1]中国船舶重工集团公司第七二三研究所,江苏扬州225101

出　　处：《舰船电子对抗》2020年第2期61-64,共4页Shipboard Electronic Countermeasure

摘　　要：目标识别是现代电子战中分析敌我的重要手段之一,对目标识别对抗系统进行多目标识别的分析研究具有重要意义。基于Mean Shift算法的原理,提出一种改进的聚类算法来进行目标识别对抗系统的多目标识别,它能够在聚类的同时诊断出复杂环境下的干扰数据点,同时相对于DBSCAN方法,能够避免数据边界点造成的聚类效果不佳,从而进行系统的多目标识别。Target recognition is one of the important means to analyze enemy and ourselves in modern electronic warfare,and it is of great significance to analyze and study the multi-target recognition of target recognition countermeasure system.Based on the principle of Mean Shift algorithm,this paper proposes an improved clustering algorithm to perform multi-target recognition of target recognition countermeasure system,which can cluster collected data and diagnose the jamming data point in complex environment.The proposed method outperforms the standard DBSCAN method,because it can avoid the bad clustering results caused by data boundary points and then performs multi-target recognition of the system.

关 键 词：多目标识别 Mean SHIFT算法 干扰数据点 

分 类 号：TN971.1[电子电信—信号与信息处理]