Proposedis the optimization of Lane Assistance Linein a vehicle speed detection system based on traffic surveillancevideos using the YOLOX algorithm and automatic homography techniques. The Lane Assistance Line plays a crucial role in calculating vehicle speed as vehicles pass two parallel lines; however, this process is often affected by environmental variables such as video resolution and lighting conditions. By utilizing YOLOX, supporting objects such as road segments, road markings, ground position, and traffic signs can be efficiently detected. The Lane Assistance Line is then projected using automatic homography to improve the accuracy of speed calculations. Experimental results show that YOLOX-l achievesthe highest processing speed (2.16 FPS), while YOLOv9-c offers an optimal balance between speed and accuracy, with an average precision of $74 \%$ and recall of $61 \%$ Despite challenges such as precision in detecting supporting objects and environmental variations, this approach enhances the accuracyof line detection and overall system performance. The YOLOX and automatic homography-based approach makes a significant contribution to the development of a more efficient and accuratevideo-based speed detection system, especially in dynamic trafficconditions.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Automatic Lane Assistance Line Detection using YOLOX in Vehicle Speed Detection System


    Beteiligte:
    Rahman, Fandisya (Autor:in) / Yuniarti, Anny (Autor:in) / Saikhu, Ahmad (Autor:in)


    Erscheinungsdatum :

    24.10.2024


    Format / Umfang :

    503942 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    YOLOX-based dynamic vehicle target detection

    Wang, Mingshen / Zhuang, Xufei / Zhao, Haitong | IEEE | 2022


    Residual YOLOX-based Ship Object Detection Method

    Liu, Ming / Zhu, Changming | IEEE | 2022


    Vehicle pedestrian detection algorithm at ferry entrance based on improved YOLOX

    Liu, Yushan / Yang, Xinyi / Liu, Weikang et al. | SPIE | 2023


    Research on YOLOX-based tire defect detection method

    Qinghua, Qi / Xiaowei, Xu / Liu, Zhan et al. | IEEE | 2022


    Lane separation line detection correcting device, lane separation line detection correcting method, and automatic driving system

    TANIGUCHI TAKUYA / KOKIDO NORITAKA / IIDA KOJI | Europäisches Patentamt | 2024

    Freier Zugriff