[1]叶靓玲,李伟达,郑力新,等.结合目标检测与特征匹配的多目标跟踪算法[J].华侨大学学报(自然科学版),2021,42(5):661-669.[doi:10.11830/ISSN.1000-5013.202105018]
　YE Liangling,LI Weida,ZHENG Lixin,et al.Multiple Object Tracking Algorithm Based on Detection and Feature Matching[J].Journal of Huaqiao University(Natural Science),2021,42(5):661-669.[doi:10.11830/ISSN.1000-5013.202105018]
点击复制

结合目标检测与特征匹配的多目标跟踪算法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第42卷
期数:
2021年第5期
页码:
661-669
栏目:
出版日期:
2021-09-20

文章信息/Info

Title:
Multiple Object Tracking Algorithm Based on Detection and Feature Matching
文章编号:
1000-5013(2021)05-0661-09
作者:
叶靓玲1;  李伟达1;  郑力新1;  曾远跃2;  黄凯2
1. 华侨大学 工学院, 福建 泉州 362021;2. 福建省特种设备检验研究院 泉州分院, 福建 泉州 36021
Author(s):
YE Liangling1;  LI Weida1;  ZHENG Lixin1;  ZENG Yuanyue2;  HUANG Kai2
1. College of Engineering, Huaqiao University, Quanzhou 362021, China; 2. Quanzhou Branch, Fujian Special Equipment Inspection and Research Institute, Quanzhou 362021, China
关键词:
多目标跟踪;  目标检测;  特征匹配;  深度学习;  YOLOv5
Keywords:
multiple object tracking;  target detection;  feature matching;  deep learning;  YOLOv5
分类号:
TP391.41
DOI:
10.11830/ISSN.1000-5013.202105018
文献标志码:
A
摘要:
针对多目标跟踪算法在遮挡频繁的场景下存在目标关联准确性低的问题,提出一种结合检测与特征匹配的多目标跟踪算法. 该算法引入检测精度较高的YOLOv5作为多目标跟踪的检测器,能够精准定位目标,有效提高跟踪精度;在面对目标间遮挡时,通过专门设计特征匹配模型提取更为细致的特征,能够有效降低跟踪时目标ID的切换次数.在MOT16数据集上对跟踪性能进行评估,结果表明:所提方法可以有效缓解目标遮挡,实现稳定跟踪.
Abstract:
Aiming at the problem that the multiple object tracking algorithm(MOT)had low accuracy of target association in frequent occlusion scenes, an MOT algorithm based on detection and feature matching is proposed in this paper. This algorithm introduces YOLOv5 with high detection accuracy as a detector for MOT, which can accurately locate the target and effectively improve the tracking accuracy. In addition, a feature matching model is specially designed when facing the goals keep out. This can extract more detailed features and effectively reduce the ID switching numbers during tracking. The tracking feature is evaluated on the MOT16 dataset, and the results show that the proposed algorithm can effectively alleviate the occlusion of the target and achieve stable tracking.

参考文献/References:

[1] XIANG Jun,XU Guohan,MA Chao,et al.End-to-end learning deep CRF models for multi-object tracking[J].IEEE Transactions on Circuits and Systems for Video Technology,2021,31(1):275-288.DOI:10.1109/TCSVT.2020.2975842.
[2] RISTANI E,TOMASI C.Features for multi-target multi-camera tracking and re-identification[C]//IEEE Conference on Computer Vision and Pattern Recognition.Salt Lake City: IEEE Press,2018:6036-6046.DOI:10.1109/CVPR.2018.00632.
[3] 顾培婷,黄德天,黄炜钦,等.抗遮挡的相关滤波目标跟踪算法[J].华侨大学学报(自然科学版),2018,39(4):611-617.DOI:10.11830/ISSN.1000-5013.201608031.
[4] 李星辰,柳晓鸣,成晓男.融合YOLO检测的多目标跟踪算法[J].计算机工程与科学,2020,42(4):665-672.DOI:10.3969/j.issn.1007-130X.2020.04.013.
[5] WANG Jiahui,GUO Yulan,TANG Xing,et al.Semi-online multiple object tracking using graphical tracklet association[J].IEEE Signal Processing Letters,2018,25(11):1725-1729.DOI:10.1109/LSP.2018.2872403.
[6] BEWLEY A,GE Zongyuan,OTT L,et al.Simple online and realtime tracking[C]//IEEE International Conference on Image Processing.Phoenix: IEEE Press,2016:3464-3468.DOI:10.1109/ICIP.2016.7533003.
[7] REDMON J,FARHADI A.YOLOv3: An incremental improvement[EB/OL].[2018-04-08] .https://arxiv.org/abs/1804.02767.
[8] WANG C Y,LIAO H Y M,YEH I H,et al.CSPNet: A new backbone that can enhance learning capability of CNN[C]//IEEE Conf-erence on Computer Vision and Pattern Recognition Workshops.Seattle: IEEE Press,2020:1571-1580.DOI:10.1109/CVPRW50498.2020.00203.
[9] ZHENG Liang,ZHANG Hengheng,SUN Shaoyan,et al.Person re-identification in the wild[C]//IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE Press,2017:3346-3355.DOI:10.1109/CVPR.2017.357.
[10] WOJKE N,BEWLEY A,PAULUS D.Simple online and realtime tracking with a deep association metric[C]//IEEE International Conference on Image Processing.Beijing: IEEE Press,2017:3645-3649.DOI:10.1109/ICIP.2017.8296962.
[11] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas: IEEE Press,2016:770-778.DOI:10.1109/CVPR.2016.90.
[12] WEINBERGER K,SAUL L.Distance metric learning for large margin nearest neighbor classification[J].Journal of Machine Learning Research,2009,10:207-244.DOI:10.5555/1577069.1577078.
[13] SCHROFF F,KALENICHENKO D,PHILBIN J.FaceNet: A unified embedding for face recognition and clustering[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Boston: IEEE Press,2015:815-823.DOI:10.1109/CVPR.2015.7298682.
[14] KUHN H W.The hungarian method for the assignment problem[M].JüNGER M,LIEBLING T M,NADDEF D,et al.50 Years of Integer Programming 1958-2008: From the Early Years to the State-of-the-Art.Heidelberg: Springer Berlin Heidelberg,2010:29-47.
[15] MILAN A,LEAL-TAIXE L,REID I,et al.MOT16: A benchmark for multi-object tracking[EB/OL].(2016-03-02)[2016-05-03] .https://arxiv.org/abs/1603.00831v2.
[16] WANG Zhongdao,ZHENG Liang,LIU Yixuan,et al.Towards real-time multi-object tracking[C]//European Conference on Computer Vision.Cham: Springer International Publishing,2020:107-122.DOI:10.1007/978-3-030-58621-8_7.
[17] REN Shaoqing,HE Kaiming,GIRSHICK R,et al.Faster R-CNN: Towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(6):1137-1149.DOI:10.1109/TPAMI.2016.2577031.
[18] WOJKE N,BEWLEY A.Deep cosine metric learning for person re-identification[C]//IEEE Winter Conference on Applications of Computer Vision.Lake Tahoe: IEEE Press,2018:748-756.DOI:10.1109/WACV.2018.00087.
[19] KALAYEH M M,BASARAN E,G?KMEN M,et al.Human semantic parsing for person re-identification[C]//IEEE Conference on Computer Vision and Pattern Recognition.Salt Lake City: IEEE Press,2018:1062-1071.DOI:10.1109/CVPR.2018.00117.
[20] DAI Zuozhuo,CHEN Mingqiang,GU Xiaodong,et al.Batch dropblock network for person re-identification and beyond[C]//IEEE International Conference on Computer Vision.Seoul: IEEE Press,2019:3690-3700.DOI:10.1109/ICCV.2019.00379.
[21] WANG Cheng,ZHANG Qian,HUANG Chang,et al.Mancs: A multi-task attentional network with curriculum sampling for person re-identification[C]//European Conference on Computer Vision.Cham: Springer International Publishing,2018:384-400.DOI:10.1007/978-3-030-01225-0_23.
[22] KIM C,LI F,REHG J M.Multi-object tracking with neural gating using bilinear LSTM[C]//European Conference on Computer Vision.Cham: Springer International Publishing,2018:208-224.DOI:10.1007/978-3-030-01237-3_13.
[23] BAE S,YOON K.Confidence-based data association and discriminative deep appearance learning for robust online multi-object tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(3):595-610.DOI:10.1109/TPAMI.2017.2691769.
[24] FU Zeyu,ANGELINI F,CHAMBERS J,et al.Multi-level cooperative fusion of GM-PHD filters for online multiple human tracking[J].IEEE Transactions on Multimedia,2019,21(9):2277-2291.DOI:10.1109/TMM.2019.2902480.
[25] LEE S,KIM M,BAE S.Learning discriminative appearance models for online multi-object tracking with appearance discriminability measures[J].IEEE Access,2018,6:67316-67328.DOI:10.1109/ACCESS.2018.2879535.
[26] BAN Y, BA S, ALAMEDA-PINEDA X,et al.Tracking multiple persons based on a variational Bayesian model[C]//European Conference on Computer Vision.Cham: Springer International Publishing,2016:52-67.DOI:10.1007/978-3-319-48881-3_5.

相似文献/References:

[1]李伟达,叶靓玲,郑力新,等.面向扶梯不安全行为的改进型深度学习检测算法[J].华侨大学学报(自然科学版),2022,43(1):119.[doi:10.11830/ISSN.1000-5013.202105059]
　LI Weida,YE Liangling,ZHENG Lixin,et al.Improved Deep Learning Detection Algorithm for Unsafe Escalator Behavior[J].Journal of Huaqiao University(Natural Science),2022,43(5):119.[doi:10.11830/ISSN.1000-5013.202105059]

备注/Memo

备注/Memo:
收稿日期: 2021-05-14
通信作者: 郑力新(1967-),男,教授,博士,主要从事运动控制、机器视觉、图像处理与模式识别的研究.E-mail:zlx@hqu.edu.cn.
基金项目: 福建省科技计划项目(2020Y0039); 福建省泉州市高层次人才创新创业项目(2020C042R)
更新日期/Last Update: 2021-09-20