Abstract In torrential rain disasters, affected areas are identified through reporting and patrolling; however, traffic monitoring remains a challenge. This study establishes a traffic anomaly detection method during heavy rain disasters. Using probe trajectory data collected during such events, the proposed method captures traffic flow characteristics using directional vectors of road inflow and outflow. The effectiveness of the proposed method was evaluated through comparisons with previously proposed methods. The results confirmed that compared with existing methods, the proposed method could more accurately detect traffic anomalies, such as U-turns and detours, and can be used to detect other traffic anomalies. Moreover, the proposed model necessitates minimal calibration effort because it requires only two parameters—time step and rate of change. The proposed method can detect various traffic anomalies that occur during a heavy rain disaster and will contribute to the early restoration of damaged areas by providing a detailed understanding of the damage situation.


    Access

    Download


    Export, share and cite



    Title :

    A Traffic Anomaly Detection Method Using Traffic Flow Vectors During Heavy Rainfall


    Additional title:

    Int. J. ITS Res.


    Contributors:


    Publication date :

    2025-04-01




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English




    A Traffic Anomaly Detection Method Using Traffic Flow Vectors During Heavy Rainfall

    Hirata, Kensuke / Kawasaki, Yosuke / Yoshida, Takahiro | Springer Verlag | 2025

    Free access

    Evaluation of the Versatility of a Traffic Anomaly Detection Method during Heavy Rainfall

    Kawasaki, Yosuke / Hirata, Kensuke / Ootake, Hiroshi | Springer Verlag | 2024

    Free access

    Evaluation of the Versatility of a Traffic Anomaly Detection Method during Heavy Rainfall

    Kawasaki, Yosuke / Hirata, Kensuke / Ootake, Hiroshi | Springer Verlag | 2024

    Free access


    Traffic Anomaly Detection Via Conditional Normalizing Flow

    Kang, Zhuangwei / Mukhopadhyay, Ayan / Gokhale, Aniruddha et al. | IEEE | 2022