In the current social environment, pedestrian road safety has become the focus of widespread attention, especially in complex traffic environments, and the effectiveness of pedestrian detection technology is particularly important. The purpose of this study is to improve the accuracy of pedestrian detection and tracking by combining YOLOv8 and ByteTrack technology to meet the detection challenges in complex environments. Experimentally, the pedestrian model is trained by YOLOv8, and the ByteTrack algorithm is used to perform multi-target tracking after the optimal model is obtained, which solves the tracking failure problem of the traditional method in the case of occlusion. In the real-time judgment process of the dangerous area, the advanced and efficient Ray Casting algorithm is adopted, and the warning mechanism is triggered when a pedestrian is detected entering the dangerous area. This study significantly improves the accuracy of pedestrian detection and warning in complex environments, provides advanced technical support for traffic safety management systems, and is of great significance for the development of smart urban transportation systems.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Detection and warning of pedestrians in dangerous areas using YOLOv8 and bytetrack multi-target tracking technology


    Beteiligte:
    Feng, Zhengang (Herausgeber:in) / Mikusova, Miroslava (Herausgeber:in) / Lin, Songyang (Autor:in)

    Kongress:

    International Conference on Smart Transportation and City Engineering (STCE 2024) ; 2024 ; Chongqing, China


    Erschienen in:

    Proc. SPIE ; 13575


    Erscheinungsdatum :

    28.04.2025





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    DANGEROUS VEHICLE WARNING SYSTEM FOR PEDESTRIANS AND BICYCLES

    FUKUSHIMA SEIYA / UENO TAKEHISA | Europäisches Patentamt | 2020

    Freier Zugriff

    DANGEROUS VEHICLE WARNING SYSTEM FOR PEDESTRIANS AND BICYCLES

    FUKUSHIMA SEIYA / UENO TAKEHISA | Europäisches Patentamt | 2020

    Freier Zugriff

    Traffic Impact Assessment System using Yolov5 and ByteTrack

    Jin Jie Ng / Kah Ong Michael Goh / Connie Tee | DOAJ | 2023

    Freier Zugriff

    Enhanced Object Detection using YOLOv8: Identifying Vehicles and Pedestrians in Urban Environments

    Bhonde, Tejas / Temare, Harshita / Dadwhal, Yogeshwar Singh | IEEE | 2024