There are many ways to track the traffic conditions on roads. With the rise of AI-based image processing technology, there has been a surge in interest in developing traffic monitoring systems that rely on camera vision data. An improved road monitoring system may be derived from the data collected in this research by placing a camera at a junction. The technique detects and classifies vehicles using a deep-learning-based methodology (YOLOv4). By comparing the positions of the vehicle with the map HD, lane-by-lane trajectories of the vehicle may be approximated. The predicted vehicle trajectories are used to calculate each lane's wait lengths and traffic volumes, broken down by travel direction. This technique was evaluated using five distinct metrics: vehicle detection rates, trajectory prediction, type of vehicle categorization, traffic volume estimate, and queue length assessment. Thousands of samples were used to assess its performance. Based on the findings, it is acceptable to assume that the system can identify vehicles with a 98% success rate and an error rate of less than 25% regarding lane-by-lane trip volume estimation and vehicle type classification. Therefore, this study's approach reveals the viability of collecting comprehensive traffic information from a camera set up at a junction. Current traffic monitoring schemes have a reasonable likelihood of being developed by this study's significant contribution, which is a strategy that combines AI with HD map approaches.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Deep Learning-Based Image Analysis to Track Vehicles


    Beteiligte:
    Prabakaran, P (Autor:in) / Jayasudha, R. (Autor:in) / Sandhu, Mukta (Autor:in) / Gupta, Sandeep (Autor:in)


    Erscheinungsdatum :

    15.03.2024


    Format / Umfang :

    602137 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Chain-track vehicles

    Daniels, J.L. | Engineering Index Backfile | 1941


    Vehicle track detection method based on deep learning

    ZHANG ZHONG / SUN GUANLONG / HUANG JUNJIE | Europäisches Patentamt | 2022

    Freier Zugriff

    Track Defect Recognition Algorithm Based on Deep Learning

    Fang, Enquan / Yang, Kunshan / Zeng, Ming | British Library Conference Proceedings | 2022


    Track Defect Recognition Algorithm Based on Deep Learning

    Fang, Enquan / Yang, Kunshan / Zeng, Ming | TIBKAT | 2022


    Track Defect Recognition Algorithm Based on Deep Learning

    Fang, Enquan / Yang, Kunshan / Zeng, Ming | Springer Verlag | 2022