现代防御技术 ›› 2020, Vol. 48 ›› Issue (1): 19-25.DOI: 10.3969/j.issn.1009-086x.2020.01.003

• 空天防御体系与武器 • 上一篇    下一篇

一种空中目标航迹聚类方法研究

梁复台1,2, 李宏权1, 孟庆文1, 蔡莉1   

  1. 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
  • 收稿日期:2019-04-10 修回日期:2019-09-04 出版日期:2020-02-20 发布日期:2021-01-20
  • 作者简介:梁复台(1983-),男,宁夏固原人。工程师,硕士生,研究方向为预警情报分析。通信地址:430019 湖北武汉市江岸区黄浦大街288号 E-mail:417516911@qq.com

An Aerial Target Track Clustering Method

LIANG Fu-tai1,2, LI Hong-quan1, MENG Qing-wen1, CAI Li1   

  1. 1. Air Force Early Warning Academy,Early Warning Intelligence Department,Hubei Wuhan 430019,China;
    2. PLA,No.31121 Troop,Jiangxi Nanchang 330000,China
  • Received:2019-04-10 Revised:2019-09-04 Online:2020-02-20 Published:2021-01-20

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。

关键词: 空中目标, 航迹聚类, 特征提取, 曲线拟合, k-means, 仿真

Abstract: 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.

Key words: aerial target, track clustering, feature extraction, curve fitting, k-means, simulation

中图分类号: