As modern cities become more intelligent, traffic flow forecasting is becoming increasingly significant in intelligent transport and urban governance. Accurate traffic forecasting can help the management and the residents to travel, and how to achieve accurate and efficient traffic forecasting is a problem that has been studied by many scholars. With the growth of artificial intelligence and the generation of large-scale traffic data, deep learning can be more effective than traditional probabilistic prediction models in traffic flow forecasting, but there are two more problems to be solved. Firstly, the use of too many levels of convolution often leads to over-fitting. Secondly, the state of traffic in a city can be impacted by external factors such as meteorological conditions, temperature, and wind speed. To solve this problem, we propose an end-to-end neural network model based on the residual network, dilated convolution, and call it STDCN. Convolution and dilated convolution are used in the model to obtain local spatial dependencies and distant spatial dependencies. We designed an external feature extractor to transform the external factors into a ID tensor using the One-Hot Encoding.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep Spatial-Temporal Network Based on Residual Networks and Dilated Convolution for Traffic Flow Prediction


    Contributors:


    Publication date :

    2022-11-11


    Size :

    589739 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Dual Dynamic Spatial-Temporal Graph Convolution Network for Traffic Prediction

    Sun, Yanfeng / Jiang, Xiangheng / Hu, Yongli et al. | IEEE | 2022


    Traffic flow prediction method based on global diffusion convolution residual network

    ZHENG KAI / YE GUANYU / LI YUANMING et al. | European Patent Office | 2020

    Free access

    Traffic Flow Prediction Using Novel Spatial-Temporal Multi-Head Attention Graph Convolution Networks

    Cheng, Yuan / Peng, Cheng / Wang, Ze et al. | Transportation Research Record | 2024



    Spatial-Temporal Position-Aware Graph Convolution Networks for Traffic Flow Forecasting

    Zhao, Yiji / Lin, Youfang / Wen, Haomin et al. | IEEE | 2023