基于均值漂移的穿墙雷达多目标跟踪

详细信息    查看全文 | 推荐本文 |

英文篇名：Mean-Shift Based Multiple Targets Tracking for TWR 作者：孔令讲 ; 陈国浩 ; 崔国龙 ; 杨晓波 英文作者：KONG Ling-jiang;CHEN Guo-hao;CUI Guo-long;YANG Xiao-bo;School of Information and Communication Engineering,University of Electronic Science and Technology of China; 关键词：均值漂移 ; 多目标跟踪 ; 图像域跟踪 ; 穿墙成像雷达 英文关键词：mean-shift;;multi-target tracking;;tracking in image domain;;TWIR 中文刊名：DKDX 英文刊名：Journal of University of Electronic Science and Technology of China 机构：电子科技大学信息与通信工程学院; 出版日期：2019-05-30 出版单位：电子科技大学学报 年：2019 期：v.48 基金：国家自然科学基金面上项目(61871080);; 中央高校基本业务费(ZYGX2014Z005) 语种：中文; 页：DKDX201903001 页数：5 CN：03 ISSN：51-1207/T 分类号：3-7

摘要

针对穿墙成像雷达中多人体目标跟踪的问题,提出了一种基于均值漂移的多目标图像域跟踪方法。首先,通过基于双轴投影的变尺度模板创建方法进行航迹起始;其次该文利用均值漂移算法进行航迹更新,并对不同幅度像素点的加权处理使得迭代过程能收敛于目标图像中心;最后,结合M/N准则,实现了对航迹的管理。通过对多组实测数据的处理可以看出,该算法具有稳健的跟踪性能。
        This paper deals with the tracking problem of multiple human targets hidden behind walls using through-wall-imaging (TWI) radar. We propose a mean-shift based multi-target tracking algorithm in image domain.Firstly, an adaptive method of target template construction based on biaxial projection is presented. Secondly, the mean-shift algorithm is used to update the trajectories, and the different amplitudes is weighted in order to make the iterative process converge to the center of the target. The M/N logic is finally considered to manage trajectories.The proposed algorithm is evaluated by the experiments.

引文

[1]van DORP P,GROEN F C A.Human walking estimation with radar[J].IEE Proceedings-Radar,Sonar and Navigation,2003,150(5):356-365.
    [2]张驰,李悦丽,周智敏.基于独立分量分析法的稀疏阵列穿墙成像雷达直达波干扰抑制[J].雷达学报,2014,3(5):524-532.ZHANG Chi,LI Yue-li,ZHOU Zhi-min.Wall clutter mitigation in through-the wall imaging radar with sparse array antenna based on independent component analysis[J].Journal of Radars,2014,3(5):524-532.
    [3]梁福来,李浩楠,祁富贵,等.UWB MIMO生物雷达多静止人体目标成像方法研究[J].雷达学报,2016,5(5):470-476.LIANG Fu-lai,LI Hao-nan,QI Fu-gui,et al.Imaging of multiple stationary humans using a UWB MIMOBio-radar[J].Journal of Radars,2016,5(5):470-476.
    [4]YUAN Xiang,KONG Ling-jiang,JIA Yong,et al.Cooperative target localization and tracking for random sparse array through-wall-radar[C]//IEEE Cie International Conference on Radar.Chengdu:IEEE,2011:1343-1345.
    [5]ROVNAKOVA J,KOCUR D.UWB radar signal processing for through wall tracking of multiple moving targets[C]//The 7th European Radar Conference.Munich:IEEE,2010:372-375.
    [6]CHEN Xiao-li,LEUNG Henry,TIAN Mao.Multitarget detection and tracking for through-the-wall radars[J].IEEETransactions on Aerospace and Electronic Systems,2014,50(2):1403-1415.
    [7]GENNARELLI G,VIVONE G,BRACA P,et al.Multiple extended target tracking for through-wall radars[J].IEEETransactions on Geoscience and Remote Sensing,2015,53(12):6482-6494.
    [8]崔国龙,孔令讲,杨建宇.步进变频穿墙成像雷达中反投影算法研究[J].电子科技大学学报,2008,37(6):864-867.CUI Guo-long,KONG Ling-jiang,YANG Jian-yu.Backprojection algorithm to stepped-frequency through-the-wall radar imaging[J].Journal of University of Electronic Science and Technology of China,2008,37(6):864-867.
    [9]LIU Jian-gang,WANG Ming-yang,KONG Ling-jiang,et al.Through-wall tracking using mean-shift algorithm[C]//IEEE Radar Conference.Seattle:IEEE,2017,2:1274-1277.
    [10]刘剑刚.穿墙雷达隐蔽目标成像跟踪方法研究[D].成都:电子科技大学,2017.LIU Jian-gang.Imaging-tracking technology for hidden targets of a through-the-wall radar[D].Chengdu:University of Electronic Science and Technology of China,2017.
    [11]COMANICIU D,RAMESH V,MEER P.Real-time tracking of non-rigid objects using mean shift[C]//IEEEConference on Computer Vision and Pattern Recognition.Flilton Head:IEEE,2002:2142.