The invention relates to the technical field of intelligent traffic systems, and discloses a sparse perception traffic scene blind area information interpolation estimation method. Acquiring traffic data collected by traffic network nodes through a sensor, and performing adaptive fusion on various feature information; carrying out dynamic time feature extraction according to the multi-modal fusion features; adding the extraction result of the dynamic time feature and the multi-modal fusion feature, and inputting the added result and the multi-modal fusion feature into a space attention module and a bidirectional time attention module to obtain a new adjacent matrix and a new dynamic transfer matrix; inputting the data to an S-fusion layer module to obtain a road network adjacency matrix with forward and backward long and short term space-time dependence; and inputting the road network adjacency matrix and the multi-modal fusion features into a graph attention gating loop module and a decoding output module to obtain interpolation estimation of nodes with missing values. According to the method, accurate capturing of complex dynamic space-time correlation characteristics among scene nodes can be enhanced, and the accuracy and robustness of scene space-time data complementation are guaranteed and improved.

    本发明涉及智慧交通系统技术领域,公开了一种稀疏感知交通场景盲区信息插补估计方法。获取交通路网节点通过传感器收集到的交通数据,对多种特征信息进行自适应融合;根据多模态融合特征进行动态时间特征提取;将动态时间特征的提取结果与多模态融合特征相加后输入到空间注意力模块和双向时间注意力模块,得到新的邻接矩阵和动态转移矩阵;再将其输入到S‑融合层模块,得到带有前后向长短期时空依赖的路网邻接矩阵;将路网邻接矩阵与多模态融合特征输入到图注意力门控循环模块和解码输出模块,得到对缺失值的节点的插补估计。本发明能够增进场景节点间的复杂动态时空关联特性准确捕获,保障提升场景时空数据补全的准确性和鲁棒性。


    Access

    Download


    Export, share and cite



    Title :

    Sparse perception traffic scene blind area information interpolation estimation method


    Additional title:

    一种稀疏感知交通场景盲区信息插补估计方法


    Contributors:
    LAN SHIYONG (author) / HOU ZHIANG (author) / QIAO YIXIN (author)

    Publication date :

    2024-11-05


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    METHOD FOR GENERATING A WIDE-AREA PERCEPTION SCENE GRAPH

    HOVIS JEFFREY GENE / SZCZERBA MICHAEL BERNHARD | European Patent Office | 2018

    Free access

    Composite scene area traffic signal optimization control method

    LI XUEFENG / LI MEITAO / ZHANG FUSHENG et al. | European Patent Office | 2024

    Free access

    Processing of LiDAR for Traffic Scene Perception of Autonomous Vehicles

    Urmila., O / Megalingam, Rajesh Kannan | IEEE | 2020


    Traffic risk perception system based on sparse point detector

    WU HONGTAO / LIU PENGYUAN / WANG JIANWEI et al. | European Patent Office | 2024

    Free access

    METHOD AND SYSTEM FOR GENERATING A WIDE-AREA PERCEPTION SCENE GRAPH

    HOVIS JEFFREY GENE / SZCZERBA MICHAEL BERNHARD | European Patent Office | 2018

    Free access