Abstract:As the number of vehicles and the volume of traffic swell in urban centers, cities have experienced a concomitant increase in traffic accidents. Proactively identifying accident-prone hotspots in urban environments holds the promise of preventing traffic mishaps, thereby curtailing the incidence of accidents and reducing property damage. This research introduces the Two-Branch Contextual Feature-Guided Converged Network (TCFGC-Net) utilizing multimodal satellite and street view data. Designed to extract global structural features from satellite imagery and dynamic continuous features from street view imagery, the model aims to improve the accuracy of detecting urban accident hotspots. For the satellite imagery branch, we propose the Contextual Feature Coupled Convolutional Neural Network (Trans-CFCCNN) designed to extract global spatial features and discern feature correlations across adjacent regions. For the street view imagery branch, we develop the Sequential Feature Recurrent Attention Network (SFRAN) to assimilate and integrate dynamic scene features captured from successive street view images. We designed the Multi-Branch Feature Adaptive Fusion Structure (MBFAF) to aggregate different branch features for accurate identification of accident hotspots. Experimental results show that the model performs well, with an overall accuracy of 93.7 %. Ablation studies confirm that relative to standalone street view and satellite branch analyses, implementing multimodal fusion enhances the model's accuracy by 12.05 % and 17.86 %, respectively. The innovative fusion structure proposed herein garners a 4.22 % increase in model accuracy, outpacing conventional feature concatenation techniques. Furthermore, the model outperforms existing deep learning models in terms of overall efficacy. Additionally, to showcase the efficacy of the proposed model structure, we utilize Class Activation Maps (CAM) to provide visual interpretability for the model. These results suggest that the dual-branch fusion model effectively decreases false alarm occurrences and directs the model's focus toward regions more pertinent to accident hotspots. Finally, the code and model used for identifying hotspots of urban traffic accidents in this study are available for access: https://github.com/gwt-ZJU/TCFGC-Net.

Identification of potential traffic accident hot spots based on accident data and GIS

Identification of black spot on traffic accidents and its spatial association analysis based on geographic information system

Analysis of Spatiotemporal Factors Affecting Traffic Safety Based on Multisource Data Fusion

Identification of Traffic Accident Hotspots using Geographical Information System (GIS)

Identification analysis model of traffic accident-prone locations based on geographical view angle

Network space analysis–based identification of road traffic accident hotspots: a case study

Determining the road traffic accident hotspots using GIS-based temporal-spatial statistical analytic techniques in Hanoi, Vietnam

Applying OHSA to Detect Road Accident Blackspots

Hotspot identification considering daily variability of traffic flow and crash record: A case study

A two-phase clustering approach for traffic accident black spots identification: integrated GIS-based processing and HDBSCAN model

Spatial Regression Analysis of Pedestrian Crashes Based on Point-of-Interest Data

Spatiotemporal Clustering and Analysis of Road Accident Hotspots by Exploiting GIS Technology and Kernel Density Estimation

Fusion of Satellite and Street View Data for Urban Traffic Accident Hotspot Identification

Urban Traffic Safety Analysis and Assessment System Based on GIS: System Design, Key Techniques and Implementation Strategy

Identification of Critical Links in Urban Road Network Based on GIS

Hotspot identification on urban arterials at the meso level

Quantitative Estimation and Analysis of Spatiotemporal Delay Effects in Expressway Traffic Accidents

Identification of Significant Factors in Fatal-Injury Highway Crashes Using Genetic Algorithm and Neural Network

GIS Based Crime Risk Analysis and Management in Cities

Identification of Traffic Accident Prone Location Based on Vehicle Speed Variability Indices

Spatial statistics and random forest approaches for traffic crash hot spot identification and prediction