Multi-Source Data Fusion Algorithm for Urban Road Traffic Accident Prevention

To meet the demand for active prevention of urban road traffic accidents, in this paper, we propose a matching and fusion algorithm based on the address and coordinates of multi-source data. First, we investigate the multi-source data and complete the data’s classification and field structure design. Next, a preprocessing procedure that includes coordinate system transformation and data localization is proposed. Subsequently, we construct the indicator system of multi-source data on traffic safety for urban roads. Finally, the independent data records are matched with the corresponding road intersections and road segments using text matching and distance discrimination techniques, respectively. The results reveal a 21.28% increase in the number of effective records in multi-source data post-preprocessing. The fused multi-source data indicator system covers comprehensive elements and has a clear hierarchical structure, providing a more reliable data foundation for research into urban road accident prevention.