基于YOLOv4与改进DeepSORT算法的车流量检测

doi:10.16046/j.cnki.issn2096-5680.2023.01.002

河北水利电力学院学报 ›› 2023, Vol. 33 ›› Issue (1): 6-11.DOI: 10.16046/j.cnki.issn2096-5680.2023.01.002

• 人工智能与机器人专题 • 上一篇下一篇

基于YOLOv4与改进DeepSORT算法的车流量检测

王继超^1,2, 张丽娟^1,2, 张春茜¹, 回振桥¹, 申耀辉¹

1.河北水利电力学院电气自动化系,河北省沧州市黄河西路49号 061001;
2.河北省高校水利自动化与信息化应用技术研发中心,河北省沧州市黄河西路49号 061001

收稿日期:2022-11-02 修回日期:2022-11-10 发布日期:2023-10-26
作者简介:王继超(1992-),男,河北沧州人,硕士,主要研究方向为深度学习、智能优化。E-mail:491366271@qq.com
基金资助:
河北省教育厅科学技术研究项目(QN2021228);河北省大学生创新创业训练计划(S202010085030);沧州市重点研发计划指导项目(204102010);沧州市重点研发计划指导项目(204102014);河北省教育厅科学技术研究项目(ZC2022018)

Traffic Flow Detection Based on YOLOv4 and Improved DeepSORT Algorithm

WANG Ji-chao^1,2, ZHANG Li-juan^1,2, ZHANG Chun-qian¹, HUI Zhen-qiao¹, Shen Yao-hui¹

1. Department of Electrical Automation, Hebei University of Water Resources and Electric Engineering, 061001, Cangzhou, Hebei, China;
2. Water Resources Automation and Informatization Application Technology and Development Center of Hebei Colleges, 061001, Cangzhou, Hebei, China

Received:2022-11-02 Revised:2022-11-10 Published:2023-10-26

摘要/Abstract

摘要： 为了提高不同时段车流量的检测率,本文在优良的目标检测模型YOLOv4基础上,将传统的DeepSORT算法进行改进,将原有的IoU变为CIoU,保留追踪信息的同时,提供了更丰富匹配策略,使得目标追踪更加稳定,在一定程度上解决了光线较暗容易丢失目标的问题。最后在车流量检测阶段将本文改进算法与SORT算法、DeepSORT算法进行对比试验。试验结果表明,在白天情况下,本文算法相较于SORT算法提高10.7%,较DeepSORT算法提高1.3%;在夜晚情况下,本文算法相较于SORT算法提高18.1%,较DeepSORT算法提高7.9%。本文改进算法在夜晚车流量检测精准,从而为环境昏暗的条件下物体目标检测与追踪提供了理论参考与方法依据。

关键词: 车流量检测, YOLOv4, DeepSORT, 目标追踪

Abstract: In order to improve the detection rate of traffic flow in different time periods, based on the excellent target detection model YOLOv4, this paper improved the traditional DeepSORT algorithm, changed the original IoU into CIoU, and provided more abundant matching strategies to make the target tracking more stable while retaining the tracking information. To a certain extent, it solves the problem that the target is easily lost when the light is dim. Finally, the improved algorithm is compared with SORT algorithm and DeepSORT algorithm in traffic flow detection stage. Experimental results show that the proposed algorithm improves by 10.7% compared with SORT algorithm and 1.3% compared with DeepSORT algorithm in daytime. At night, the proposed algorithm improves 18.1% compared with SORT algorithm and 7.9% compared with DeepSORT algorithm. In this paper, the improved algorithm is accurate in detecting traffic flow at night, which provides theoretical reference and method basis for object detection and tracking in dim environment.

Key words: traffic flow detection, YOLOv4, DeepSORT, target tracking

中图分类号:

王继超, 张丽娟, 张春茜, 回振桥, 申耀辉. 基于YOLOv4与改进DeepSORT算法的车流量检测[J]. 河北水利电力学院学报, 2023, 33(1): 6-11.

WANG Ji-chao, ZHANG Li-juan, ZHANG Chun-qian, HUI Zhen-qiao, Shen Yao-hui. Traffic Flow Detection Based on YOLOv4 and Improved DeepSORT Algorithm[J]. Journal of Hebei University of Water Resources and Electric Engineering, 2023, 33(1): 6-11.

参考文献

[1] 储泽楠,韩毅,宋倍倍.基于嵌入式GPU的车流量检测系统[J].河南科技,2020(5):29-31.
[2] 王继超,张丽娟,尤田,等.人脸识别的智能防疫系统设计[J].河北水利电力学院学报,2021,31(4):55-59.
[3] 陈佳倩,金晅宏,王文远,等.基于YOLOv3和DeepSORT的车流量检测[J].计量学报,2021,42(6):718-723.
[4] 王继超,李擎,崔家瑞,等.一种改进的人工蜂群算法—粒子蜂群算法[J].工程科学学报,2018,40(7):871-881.
[5] Girshick R,Donahue J,Darrell T,et al.Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,Columbus,2014:580-587.
[6] Girshick R.Fast R-CNN[C]//Proceedings of IEEE International Conference on Computer Vision. Santiago,Chile:IEEE,2015:1440-1448.
[7] Ren S,He K,Girshick R,et al.Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017,39(6):1137-1149.
[8] 张文,杨雅姿,黄驰,等.一种基于YOLOV4 Tiny的目标检测算法[J].电脑与信息技术,2022,30(2):33-37.
[9] LIU W,Anguelov D,Erhan D,et al.SSD:Single shot multibox detector[C]//European Conference on Computer Vision. Amsterdam,The Netherlands:Springer,2016:21-37.
[10] Redmon J,Divvala S,Girshick R,et al.You only look once:Unified,real-time object detection with region proposal networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017,39(6):1137-1149.
[11] Redmon J,Farhadi A.YOLO9000:Better,Faster,Stronger[C]//IEEE Conference on Computer Vision & Pattern Recognition,2017:6517-6525.
[12] Redmon J, FarhadiA. YOL. Ov3: An Incremental Improvement[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, USA, 2018:1-4
[13] BOCHKOVSKIY A, WANG C, LIAO H. YOL Ov4: Optimal Speed and Accuracy of Object Detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, 2020:392-401.
[14] 孙东来,王继超,陈科,等.基于Ghost-YOLOv3-2算法的2尺度猪目标检测[J].江苏农业科学,2022,50(7):189-196.
[15] 张慧,王坤峰,王飞跃.深度学习在目标视觉检测中的应用进展与展望[J].自动化学报,2017,43(8):1289-1305.
[16] 睢丙东,张湃,王晓君.一种改进YOLOv3的手势识别算法[J].河北科技大学学报,2021,42(1):22-29.
[17] 解耘宇.基于扩展卡尔曼滤波的单目视觉轨迹跟踪方法的研究[D].北京:华北电力大学,2017.
[18] Bewley A,Ge Z Y,Ott L,et al. Simple online and realtime tracking[C]//2016 IEEE International Conference on Image Processing,2016.
[19] Balajiv,John W R,Pritam C.DeepSORT:deep convolutionalnetworksforSORTinghaploidmaizeseeds [J].BMCBioinformatics,2018,19(9):85-93.
[20] 陈紫强,张雅琼.一种基于YOLOv4的改进DeepSORT目标跟踪算法[J].桂林电子科技大学学报,2021,41(2):140-145.

基于YOLOv4与改进DeepSORT算法的车流量检测

Traffic Flow Detection Based on YOLOv4 and Improved DeepSORT Algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 5

编辑推荐

Metrics

[1]	刘炳华, 郭忠印, 田泽宇, 吴琼. 双车道公路下坡弯道路段货车偏移行为仿真研究[J]. 河北水利电力学院学报, 2021, 31(4): 37-43.
[2]	冯雷. 基于仿真的高速公路施工区交通分流分析[J]. 河北水利电力学院学报, 2021, 31(3): 13-17.
[3]	郝放, 叶全喜, 崔立长. 中小城市静态交通规划建设及管理策略探究[J]. 河北水利电力学院学报, 2020, 30(4): 66-71.
[4]	韩雪, 王笛. 城市居民停车共享泊位选择行为分析[J]. 河北水利电力学院学报, 2019, 29(3): 57-60.
[5]	路坤, 张希庆, 张君纬. 关于《道路交通标志和标线》(GB5768.2-2009)心理学理论基础的几点讨论和研究[J]. 河北水利电力学院学报, 2019, 29(2): 28-32.