During the driving process, it is common for drivers to misjudge the traffic signs due to extreme weather, poor lighting conditions, and damaged traffic signs, leading to serious traffic accidents. A traffic sign recognition method based on YOLOv5 algorithm was proposed to solve these issues. Utilizing the coordinate attention mechanism to reduce complex background interference, increase the model’s attention to important features, and improve the loss function to reduce the degree of freedom of the prediction box during the convergence process, thereby improving detection performance. Experimental results show that the proposed method achieves significant performance improvement for traffic sign detection under different conditions.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Research on Traffic Target Recognition Algorithm Based on Deep Learning


    Beteiligte:
    Li, Dan (Autor:in) / Dang, Xiangying (Autor:in) / Shi, Hanqin (Autor:in) / Li, Teng (Autor:in)


    Erscheinungsdatum :

    21.10.2023


    Format / Umfang :

    983272 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Traffic signs recognition with deep learning

    Yasmina, Djebbara / Karima, Rebai / Ouahiba, Azouaoui | IEEE | 2018


    Traffic Signal Recognition System Using Deep Learning

    Nuthakki, Ramesh / Aameen, Abdullah / Kumar, Nawnit et al. | IEEE | 2023


    Traffic Sign Recognition Algorithm Model Based on Machine Learning

    Li, Hui / Feng, Jun / Liu, Jialing et al. | Springer Verlag | 2020