一种空中目标航迹聚类方法研究  

An Aerial Target Track Clustering Method

作　　者：梁复台 李宏权[1] 孟庆文[1] 蔡莉[1] LIANG Fu-tai;LI Hong-quan;MENG Qing-wen;CAI Li(Air Force Early Warning Academy,Early Warning Intelligence Department,Hubei Wuhan 430019,China;PLA,No.31121 Troop,Jiangxi Nanchang 330000,China)

机构地区：[1]空军预警学院预警情报系,湖北武汉430019 [2]中国人民解放军31121部队,江西南昌330000

出　　处：《现代防御技术》2020年第1期19-25,共7页Modern Defence Technology

摘　　要：针对目前传统航迹聚类方法的不足,提出一种空中目标航迹聚类方法。首先提出一种航迹自适应拟合算法,对空中目标航迹进行拟合提取航迹特征;然后,在所提取的航迹特征基础上提出基于k-means算法的航迹聚类方法,通过对航迹特征点及拟合曲线的聚类,将同一类航迹聚类形成相似航迹簇;最后通过实验仿真,验证了该聚类方法的有效性。Aiming at the shortcomings of the traditional track clustering methods,an aerial target track clustering method is proposed.Firstly,an adaptive track fitting algorithm is proposed to extract the track features of the air target.Then,based on the extracted track features,a track clustering method based on k-means algorithm is proposed.By clustering the track feature points and fitting curve coefficients,the same kind of tracks are clustered to form similar track clusters.Finally,the effectiveness of the clustering method is verified by simulations.

关 键 词：空中目标 航迹聚类 特征提取 曲线拟合 K-MEANS 仿真 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论] E954[自动化与计算机技术—计算机科学与技术]