The invention provides an urban road network traffic accident risk prediction method and system based on spatio-temporal data fusion, and the method comprises the following steps: carrying out the feature extraction of three types of spatial data: accident latitude and longitude data, satellite image data and POI data, and constructing a distance adjacency matrix, a similar adjacency matrix and a POI feature matrix; feature extraction is carried out on three types of time series data of a road network structure, traffic flow data and meteorological data, and a risk feature matrix, a traffic flow feature matrix and a weather feature matrix are constructed respectively; obtaining a fusion adjacent matrix and a fusion feature matrix; inputting the adjacent matrix and the fusion feature matrix into a GCN for processing, inputting GCN output information into a GRU for processing, and performing adaptive weight dynamic allocation on a GRU output result through an attention mechanism; and the attention mechanism output feature matrix outputs a prediction result through an activation function. According to the invention, traffic accident risk prediction of the whole urban road network can be realized.
本发明提供了一种基于时空数据融合的城市路网交通事故风险预测方法及系统,包括如下步骤:对事故经纬度数据、卫星图像数据和POI数据这三类空间数据进行特征提取,并分别构建距离邻接矩阵、相似邻接矩阵和POI特征矩阵;对道路网络结构、交通流数据和气象数据这三类时序数据进行特征提取,并分别构建风险特征矩阵、交通流量特征矩阵和天气特征矩阵;获得融合邻接矩阵和融合特征矩阵;将邻接矩阵和融合特征矩阵输入GCN处理,GCN输出信息输入GRU处理,通过注意力机制对GRU输出结果进行自适应权重动态分配;注意力机制输出特征矩阵通过激活函数输出预测结果。本发明能够实现整个城市路网的交通事故风险预测。
Urban road network traffic accident risk prediction method and system based on spatio-temporal data fusion
一种基于时空数据融合的城市路网交通事故风险预测方法及系统
2024-05-28
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06V |
Urban traffic prediction method and system based on spatio-temporal data flow fusion analysis
European Patent Office | 2020
|Urban road network traffic accident risk studying, judging and visualizing method
European Patent Office | 2025
|Real-time road traffic prediction with spatio-temporal correlations
Online Contents | 2011
|