With the continuous deepening of research in the field of computer vision, related technologies have gradually been applied in the field of unmanned driving. This paper first briefly introduces the principles and algorithm implementation ideas of algorithms for multi-target detection and tracking in unmanned street scenes in complex scenes. Based on analyzing and comparing the advantages and disadvantages of these algorithms, we propose our improved YOLOV5 and the target detection and tracking algorithm of DeepSORT. Moreover, the theoretical analysis and implementation details of the theory’s image enhancement, target detection and tracking are discussed. Finally, the algorithm is tested. Test results show that our algorithm based on improved YOLOV5 and DeepSORT performs better than other algorithms in most quantitative evaluation metrics.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Multi-objective Detection and Tracking Algorithm for Unmanned Street View Based on Improved YOLOv5 and DeepSORT under Complex Scenes


    Beteiligte:
    Zhang, Chi (Autor:in) / Zhang, Hanyun (Autor:in) / Zhang, Chenyang (Autor:in)


    Erscheinungsdatum :

    12.10.2022


    Format / Umfang :

    1551682 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Traffic Flow Detection Based on Improved YOLOv5 and DeepSORT

    Lin, Xujie / Sun, Huadong / Chen, Ziyang | IEEE | 2024


    Real-Time Detection and Tracking of Express Parcels Based on Improved YOLOv5+DeepSORT

    Liu, Qikong / Wu, Jian / Yin, Lei et al. | Springer Verlag | 2023




    Smart Junctions: Yolov5, DeepSORT, and Direction Analysis for Traffic Enhancement

    U, Syed Abudhagir / Koushik, Katakam / Reddy, Karkala Vikas et al. | IEEE | 2024