计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (11A): 220900164-9.doi: 10.11896/jsjkx.220900164

• 大数据&数据科学 • 上一篇    下一篇

基于深度时间对比的中断航迹关联方法

侯海轮, 雷艺, 魏博, 樊玉琦   

  1. 合肥工业大学计算机与信息学院 合肥 230601
  • 发布日期:2023-11-09
  • 通讯作者: 樊玉琦(yuqi.fan@hfut.edu.cn)
  • 作者简介:(hailunhou@mail.hfut.edu.cn)
  • 基金资助:
    安徽省重点研发计划(201904a07020030);国家自然科学基金青年基金(62002097)

Track Segment Association Based on Deep Temporal Contrasting

HOU Hailun, LEI Yi, WEI Bo, FAN Yuqi   

  1. School of Computer Science and Information Engineering,Hefei University of Technology,Hefei 230601,China
  • Published:2023-11-09
  • About author:HOU Hailun,born in 1998,postgra-duate.His main research interests include spatiotemporal data mining and so on.
    FAN Yuqi,born in 1976,Ph.D,associate professor.His main research interests include blockchain,spatiotemporal data mining and so on.
  • Supported by:
    Key Research and Development Program of Anhui Province,China(201904a07020030) and National Natural Science Foundation of China(62002097).

PDF (PC)

597

摘要： 雷达在跟踪飞行目标时,常发生跟踪中断的现象。深度学习方法具有强大的学习能力,已被逐渐用于解决中断航迹关联问题。但已有基于深度学习的中断航迹关联方法未能充分考虑新老航迹特征之间的相似性,关联性能仍有待提升。因此提出一种基于深度时间对比的中断航迹关联算法(Track Segment Association based on Deep Temporal Contrasting,TSADTC),包括航迹特征提取模块、时间对比模块、航迹特征对比模块和分类器模块。航迹特征提取模块利用双向长短期记忆网络和编码器-解码器分别对新老航迹进行特征提取;时间对比模块分别使用一条航迹的特征预测另外一条航迹;航迹特征对比模块计算两条航迹的特征差别,并将此差别输入分类器中,以计算两条航迹关联的概率。算法将关联概率最大的航迹对作为关联航迹对。仿真实验表明,TSADTC算法能够有效提高中断航迹关联的正确关联率、错误关联率和航迹漏关联率性能。

关键词: 时间对比, 中断航迹关联, 编码器-解码器, 双向长短期记忆网络

Abstract: The radar’s tracking of a flying target is often interrupted,which seriously affects the perception of the airfield situation.Deep learning has powerful learning capabilities and has been gradually used to solve the problem of interrupted track asso-ciation.However,the existing deep learning-based interrupted track association methods fail to fully consider the similarity between the old and new track features,hence the association performance needs to be improved.Therefore,this paper proposes a track segment association algorithm based on deep temporal contrasting(TSADTC),which includes a track feature extraction mo-dule,a time comparison module,a track feature comparison module and a classifier module.The track feature extraction module uses the bidirectional LSTM(Bi-LSTM) and the encoder-decoder to extract the features of the new and the old tracks,respectively.In the time comparison module,the features of a track are used to predict the other track,so that the features of the two tracks of the same target have high similarity.The track feature comparison module calculates the feature difference of the two tracks,which is fed into the classifier to decide the association probability of the two tracks.The track pair with the largest association probability is set as the associated tracks.Experimental results show that the proposed algorithm TSADTC can effectively improve the performance of correct association rate,false association rate and missing track association rate of interrupted track association.

Key words: Temporal contrasting, Track segment association, Encoder-Decoder, Bi-LSTM

中图分类号: 

  • TP391
[1]ZHANG S P,LIU H M L,YU K,et al.A Robust Filtering Al-gorithm for Radar Target Tracking[J].Air&Space Defense,2020,3(1):73-79.
[2]BO J T,WANG G H,YU H B,et al.Hypersonic Target Detection and Correlation Algorithm under the Condition of Large Time-Space Difference[J].Modern Defence Technology,2021,49(5):56-64.
[3]WU S Y,LI D S,XUE Q T X,et al.Research on PHD Filter Based Track-Before-Detect Algorithm of Radar Fluctuating Targets[J].Acta Electronica Sinica,2022,50(3):691-702.
[4]WANG S L,BI D P,RUAN H L R,et al.Cognitive Radar Maneuvering Target Tracking Algorithm Based on Information Entropy Criterion[J].Acta Electronica Sinica,2019,47(6):1277-1284.
[5]HAN W,CHEN C S,LI Z H.A Method of Track Segment Association in Doppler Blind Zone[J].Radar Science and Technology,2020,18(3):239-246.
[6]WANG Y,HAN W.A New Plot-track Correlation Method ofMultiple-target on the Condition of Doppler Blind Zone[J].Fire Control & Command Control,2020,45(5):58-63.
[7]RAGHU J,SRIHARI P,THARMARASA R,et al.Comprehensive track segment association for improved track continuity[J].IEEE Transactions on Aerospace and Electronic Systems,2018,54(5):2463-2480.
[8]CHEN Y L,TONG Q,SHEN T T,et al.Short-term Trajectory Prediction Model Based on Attention Mechanism and Generative Adversarial Network[J].Journal of Computer Applications,2022:3292-3299.
[9]YANG K,DANINO M,BAR-SHALOM Y,et al.Track segment association in target selection for interdiction using a single passive sensor[J].IEEE Transactions on Aerospace and Electronic Systems,2021,57(5):3268-3278.
[10]XIONG W,XU P,CUI Y,et al.Track Segment Association via track graph representation learning[J].IET Radar,Sonar & Navigation,2021,15(11):1458-1471.
[11]MUCCI R,ARNOLD J,BAR-SHALOM Y.Track segment association with a distributed field of sensors[J].The Journal of the Acoustical Society of America,1985,78(4):1317-1324.
[12]YEOM S W,KIRUBARAJAN T,BAR-SHALOM Y.Improving track continuity using track segment association[C]//2003 IEEE Aerospace Conference Proceedings(Cat.No.03TH8652),2003:4_1925-4_1941.
[13]CASTNNÓN G,FINN L.Multi-target tracklet stitchingthrough network flows[C]//2011 Aerospace Conference.2011:1-7.
[14]QI L,WANG H P,XIONG W,et al.Track Segment Association Algorithm Based on Multiple-hypothesis Models with Priori Information[J].Systems Engineering and Electronics,2015,37(4):732-739.
[15]WANG Z H,HAN H.Tracking Segment Association Algorithm for Airborne Radar[J].Modern Radar,2019,41(7):61-64.
[16]ZHOUX P,LI J J,ZHAO X L Z,et al.Airborne Radar Track Segment Association Based on Hungarian Algorithm[J].Mo-dern Radar,2021,43(6):42-48.
[17]ZHANG W,HU M,DU J.An end-to-end framework for flight trajectory data analysis based on deep autoencoder network[J].Aerospace Science and Technology,2022,127:107726.
[18]WANG L,TANG J,LIAO Q.A study on radar target detection based on deep neural networks[J].IEEE Sensors Letters,2019,3(3):1-4.
[19]LIU J,WANG Z,XU M.DeepMTT:A deep learning maneuve-ring target-tracking algorithm based on bidirectional LSTM network[J].Information Fusion,2020,53:289-304.
[20]RÜTTGERS M,LEE S,JEON S,et al.Prediction of a typhoon track using a generative adversarial network and satellite images[J].Scientific Reports,2019,9(1):1-15.
[21]XIONG W,XU P,CUI Y,et al.Track Segment AssociationWith Dual Contrast Neural Network[J].IEEE Transactions on Aerospace and Electronic Systems,2021,58(1):247-261.
[22]HAN L L,XIU J J,DONG K,et al.A Space Target Intermittent Track Association Algorithm Based on Trajectory Prediction[J].Electronics Optics & Control,2019,26(12):100-104.
[23]QI L,WANG H P,LIU Y.Track Segment Association Algorithm Based on Statistical Binary Thresholds[J].Journal of Radars,2015,4(3):301-308.
[24]CAO Y,CAO J,ZHOU Z.Track Segment Association Method Based on Bidirectional Track Prediction and Fuzzy Analysis[J/OL].Aerospace,2022,9(5):274.https://doi.org/10.3390/aerospace9050274.
[25]SUN J,WANG N,ZHANG Z.Track Segment Association ofManeuvering Target Based on Expectation Maximization[C]//2018 11th International Congress on Image and Signal Proces-sing,BioMedical Engineering and Informatics(CISP-BMEI).2018:1-5.
[26]LIN L,BAR-SHALOM Y,KIRUBARAJAN T.New assign-ment-based data association for tracking move-stop-move targets[J].IEEE Transactions on Aerospace and Electronic Systems,2004,40(2):714-725.
[27]GUO Y,ZHU J,ZHOU S,et al.A joint model and data driven track segment association algorithm for manoeuvring target tracking[J].IET Radar,Sonar & Navigation,2022,16(10):1670-1680.
[28]MORI S,CHONG C Y.Performance analysis of graph-basedtrack stitching[C]//Proceedings of the 16th International Conference on Information Fusion.2013:196-203.
[29]VAN DER MERWE L J,DE VILLIERS J P.Track-stitchingusing graphical models and message passing[C]//Proceedings of the 16th International Conference on Information Fusion.2013:758-765.
[30]CHAI L,KONG L,LI S,et al.The multiple model multi-Ber-noulli filter based track-before-detect using a likelihood based adaptive birth distribution[J].Signal Processing,2020,171:107501.
[31]HAN W,TANG Z Y,ZHU Z B.A Method of Tracking Targets in Doppler Blind Zone Based on Multiple-hypothesis Motion Models[J].Journal of Electronics & Information Technology,2013,35(3):516-523.
[32]MAO Y F,ZHANG D L Z,WANG L W.Fuzzy Theory and Evidence Combination Based Track Segment Association for Ballistic Missile in Unpowered Phase[J].Electronics Optics & Control,2017,24(2):47-50.
[33]YU J G,LIU M,CHEN J H.Ballistic Target Track Segments Association and Optimization[J].Acta Aeronautica et Astronautica Sinica,2011,32(10):1897-1904.
[34]XU P L,CUI Y Q,XIONG W,et al.Generative Track Segment Consecutive Association Method[J].Systems Engineering and Electronics,2022,44(5):1543-1552.
[35]QIN W,TANG J,LAO S.DeepFR:A trajectory prediction mo-del based on deep feature representation[J].Information Sciences,2022,604:226-248.
[36]LI S,JIN X,XUAN Y,et al.Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting[J/OL].Advances in Neural Information Processing Systems,2019,32.https://proceedings.neurips.cc/paper/2019/hash/6775a0635c302542da2c32aa19d86be0-Abstract.html.
[37]XIU J J,ZHANG J Y,DONG K D.Research on Space-targetTrack Segments Association Algorithm[J].Journal of Ballistics,2020:85-90.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发