In intelligent transportation systems, pedestrian intentions at intersections must be classified to enhance pedestrian safety and optimize traffic flow. Accurately predicting pedestrian actions can significantly reduce traffic accidents and improve overall traffic management efficiency. An improved YOLOv5 model is proposed to efficiently and accurately identify pedestrian intentions. The model combines the Multi-scale Diluted Attention (MSDA) mechanism, which extends attention to different scales to capture the subtle movements of pedestrians, and the GhostNet lightweight module, which reduces the number of computational parameters and makes the model suitable for real-time applications. In addition, a new loss function, ACFloss, which combines adaptive weight-focusing loss based on contextual relationships and spatio-temporal consistency loss, is designed to understand the behavioral intent of actions better. Experimental results show that the improved YOLOv5 model achieves a classification accuracy of 93.7%, 4.5% higher than the benchmark, and the inference speed is improved by 35%. This improvement provides a reliable solution for intelligent transportation systems in complex intersection scenarios.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep-learning-based algorithm for classifying pedestrian behavior at crosswalks


    Contributors:
    Liu, Bin (editor) / Leng, Lu (editor) / Li, Menghao (author) / Liu, Miao (author)

    Conference:

    Third International Conference on Image Processing, Object Detection, and Tracking (IPODT 2024) ; 2024 ; Nanjing, China


    Published in:

    Proc. SPIE ; 13396


    Publication date :

    2024-10-24





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Pedestrian behavior at signal-controlled crosswalks

    Fugger,T.F. / Randles,B.C. / Wobrock,J.L. et al. | Automotive engineering | 2001


    Pedestrian Crossing Behavior at Signalized Crosswalks

    Xie, S. Q. / Wong, S. C. / Ng, Tsz Man et al. | ASCE | 2017


    Pedestrian Behavior at Signal-Controlled Crosswalks

    Stein, Anthony C. / Randles, Bryan C. / Wobrock, Jesse L. et al. | SAE Technical Papers | 2001


    Feature Relevance Estimation for Learning Pedestrian Behavior at Crosswalks

    Volz, Benjamin / Mielenz, Holger / Agamennoni, Gabriel et al. | IEEE | 2015


    Pedestrian Crosswalks at Midblock Locations

    Lu, George (Xiao-Zhao) / Noyce, David A. | Transportation Research Record | 2009