计算机科学 ›› 2023, Vol. 50 ›› Issue (9): 176-183.doi: 10.11896/jsjkx.220900004
王雒, 李飚, 傅瑞罡
WANG Luo, LI Biao, FU Ruigang
摘要： 红外目标智能检测跟踪技术研究一直是同领域中的热点问题,尤其是在精确制导、海面监视和天空预警等方面。针对红外地面多目标跟踪场景中,由地面杂波干扰、多目标遮挡干扰、平台晃动等复杂场景造成的跟踪精度降低等问题,提出了一种基于改进ByteTrack算法的红外地面多目标跟踪方法。首先引用一种自适应调制噪声尺度的卡尔曼滤波器,缓解低质量检测对vanilla卡尔曼滤波器的影响;其次引入增强相关系数最大化算法对帧间图像进行配准,来补偿平台晃动产生的影响;然后增加了基于长短期记忆网络的运动模型,减小了卡尔曼滤波在非线性运动状态中产生的预测误差;最后引入连接模型和高斯平滑算法这两种轻量级离线算法来完善跟踪结果。在红外地面多目标数据集上进行了实验,结果表明,与Sort和Deepsort算法相比,改进算法的MOTA值分别提升了8.3%和10.2%,IDF1值分别提升了6.5%和5.6%。与同类算法相比,改进算法表现出了更好的有效性,在红外目标智能检测跟踪场景中会有较大应用。
中图分类号:
| [1]LIU C.Multi-object Fault-tolerant Tracking and Trajectory Pre-diction for Unmanned Vehicles [D].Hangzhou:Zhejiang University,2019. [2]FU R G,FAN H Q,ZHU Y F,et al.A dataset for infrared time-sensitive target detection and tracking for air-ground application [J].China Scientific Data,2022,7(2):203-218. [3]ZHANG Y F,SUN P Z,JIANG Y,et al.Byte-Track:Multi-object Tracking by Associating Every Detection Box [EB/OL].https://arxiv.org/pdf/2110.06864.pdf,2022. [4]DU Y H,WAN J F,ZHAO Y Y,et al.A Com-prehensiveFramework for MCMOT With Global Information and Optimizing Strategies in VisDrone 2021[C]//Proceedings of the IEEE/CVF Inter-national Conference on Computer Vision.2021:2809-2819. [5]ESS A,LEIBE B,SCHINDLER K,et al.A mobile vision system for robust multi-person tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2008:1-8. [6]DU Y H,SONG Y,YANG B,et al.StrongSORT:Make DeepSORT Great Again [EB/OL].https://arxiv.org/pdf/2202.13514.pdf,2022. [7]KIM C,LI F,CIPTADI A,et al.Multiple hypothesis trackingrevisited[C]//IEEE International Conference on Computer Vision.2015:4696-4704. [8]REID D B.An algorithm for tracking multiple targets[J].IEEE Transactions on Automatic Control,1979,24(6):843-854. [9]YOON J H,YANG M H,LIM J,et al.Bayesian multi-object tracking using motion context from multiple objects[C]//2015 IEEE Winter Conference on Applications of Computer Vision.2015:33-40. [10]BEWLEY A,GE Z Y,OTT L,et al.Simple Online and Realtime Tracking[C]//2016 IEEE International Conference on Image Processing(ICIP).2016:3464-3468. [11]WOJKE N,BEWLEY A,PAULUS D,et al.Sim-ple online and real-time tracking with a deep association metric[C]//2017 IEEE International Conference on Image Processing(ICIP).2017:3645-3449. [12]CHEN L,AI H Z,ZHUANG Z J,et al.Real-Time Multiple People Tracking with Deeply Learned Candidate Selection and Person Re-Identification [EB/OL].https://arxiv.org/pdf/1809.04427.pdf,2018. [13]WANG Z D,ZHENG L,LIU Y X,et al.Towards real-timemulti-object tracking [EB/OL].https://arxiv.org/pdf/1909.12605.pdf,2018. [14]ZHANG Y F,WANG C Y,WANG X G,et al.FairMOT:On the fairness of detection and re-identification in multiple object tracking[EB/OL].https://arxiv.org/pdf/2004.01888.pdf. [15]ZHOU X Y,KOLTUN V,KRAHENBUHL P,et al.Tracking Objects as Points[EB/OL].https://arxiv.org/pdf/2004.01177.pdf,2020. [16]WANG L X,ZENG D,ZHU X Q.Space Based Infrared Small Target Detection and Tracking System [J].Industry Control Computer,2022,35(5):39-41. [17]SONG Z Z,YANG J W,ZHANG D F,et al.Real-time infrared multi-class multi-target anchor-free tracking network[J].Systems Engineering and Electronics,2022,44(2):401-409. [18]YANG B,LIN S Z,LU X F,et al.Multiple aerial infrared target tracking method based on multi-feature fusion and hierarchical data association [J].Journal of Computer Applications,2022,40(10):3075-3080. [19]XIE X F,LIU H J,ZHANG L J,et al.Infrared Ship TargetRecognition Fused with Lightweight YOLOv4 and KCF Algorithm[J].Journal of Ordnance Equipment Engineering,2021,42(6):175-182. [20]YANG W,YAO T,GENG X T,et al.Research on application ofinfrared target recognition technology of air defense weapon based on deep learning[J].Journal of Ordnance Equipment Engineering,2022,43(1):125-129. [21]ZHENG G,LIU S T,WANG F,et al.YOLOX:Exceeding YOLO Series in 2021 [EB/OL].https://arxiv.org/pdf/2107.08430.pdf,2021. |
|
||
首页