Traffic forecasting plays an important role in urban planning. Deep learning methods outperform traditional traffic flow forecasting models because of their ability to capture spatiotemporal characteristics of traffic conditions. However, these methods require high-quality historical traffic data, which can be both difficult to acquire and non-comprehensive, making it hard to predict traffic flows at the city scale. To resolve this problem, we implemented a deep learning method, SceneGCN, to forecast traffic speed at the city scale. The model involves two steps: firstly, scene features are extracted from Google Street View (GSV) images for each road segment using pretrained Resnet18 models. Then, the extracted features are entered into a graph convolutional neural network to predict traffic speed at different hours of the day. Our results show that the accuracy of the model can reach up to 86.5% and the Resnet18 model pretrained by Places365 is the best choice to extract scene features for traffic forecasting tasks. Finally, we conclude that the proposed model can predict traffic speed efficiently at the city scale and GSV images have the potential to capture information about human activities.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Forecasting Traffic Speed during Daytime from Google Street View Images using Deep Learning


    Weitere Titelangaben:

    Transportation Research Record: Journal of the Transportation Research Board


    Beteiligte:
    Jiao, Junfeng (Autor:in) / Wang, Huihai (Autor:in)


    Erscheinungsdatum :

    2023-05-05




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Potential Accuracy of Traffic Signs' Positions Extracted From Google Street View

    Yan, Wai Yeung / Shaker, Ahmed / Easa, Said | IEEE | 2013



    Google Street View is watching you

    Jahn, David | Online Contents | 2009


    Google Street View Images Support the Development of Vision-Based Driver Assistance Systems

    Salmen, J. / Houben, S. / Schlipsing, M. et al. | British Library Conference Proceedings | 2012


    Google Street View images support the development of vision-based driver assistance systems

    Salmen, Jan / Houben, Sebastian / Schlipsing, Marc | IEEE | 2012