The invention relates to a traffic flow prediction method based on a tensor decomposition reconstruction fusion image. The method comprises the following steps: S1, constructing a plurality of space-time diagrams into a space-time fusion image adjacency matrix; s2, arranging the spatio-temporal fusion image adjacent matrixes into a spatio-temporal tensor fusion image, extracting spatio-temporal tensor fusion image block matrix correlation, expanding a core tensor to obtain a reconstructed spatio-temporal fusion image adjacent matrix, and inputting the reconstructed spatio-temporal fusion image adjacent matrix into an image convolution neural network to construct a spatio-temporal tensor image convolution module; and S3, extracting global space-time correlation by adopting an expansion convolution module, and parallelizing the space-time tensor graph convolution module and the expansion convolution module. According to the method, through a space-time tensor graph convolution module for learning local space-time correlation and an expansion convolution module for learning global correlation, the comprehensive space-time dependency relationship of the road network can be aggregated and learned, so that the precision of traffic flow prediction is improved.

    本发明的一种基于张量分解重构融合图的交通流预测方法,步骤为:S1,将多个时空图构建成时空融合图邻接矩阵;S2,将时空融合图邻接矩阵排列成时空张量融合图,提取时空张量融合图块矩阵相关性,并将核心张量展开,获得重构时空融合图邻接矩阵,将其输入图卷积神经网络中,以构建时空张量图卷积模块;S3,采用膨胀卷积模块提取全局时空相关性,并将时空张量图卷积模块和膨胀卷积模块并行。该方法通过局部时空相关性学习的时空张量图卷积模块和学习全局相关性的膨胀卷积模块,使得道路网络的综合时空依赖关系能够被聚合和学习,从而提高交通流预测的精度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic flow prediction method based on tensor decomposition and reconstruction fusion image


    Weitere Titelangaben:

    一种基于张量分解重构融合图的交通流预测方法


    Beteiligte:
    LI QIN (Autor:in) / YANG XUAN (Autor:in) / ZHENG ZUOCAI (Autor:in) / CHEN XUANJUN (Autor:in) / HE DEQIANG (Autor:in) / XU PAI (Autor:in) / OU BINGGUANG (Autor:in)

    Erscheinungsdatum :

    10.09.2024


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Traffic flow prediction method and system under data loss based on tensor decomposition

    WANG CHAO / ZHANG YUNFAN / WU WEILING et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on fusion delay reconstruction and GRU-SVR

    LEI YUHANG / LEI JINGSHENG / TANG XIAOLAN | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow data completion method and system based on tensor chain decomposition

    ZHU CHENLU / YIN PU / YANG ZECAN et al. | Europäisches Patentamt | 2023

    Freier Zugriff


    Traffic flow prediction method based on feature reconstruction error

    YU ZHENGXU / CAI DENG / WANG PENGFEI et al. | Europäisches Patentamt | 2020

    Freier Zugriff