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.

Traffic density estimation via a multi-level feature fusion network

Turning traffic surveillance cameras into intelligent sensors for traffic density estimation

Multi-Dimensional Traffic Congestion Detection Based on Fusion of Visual Features and Convolutional Neural Network

Crowd density estimation based on multi scale features fusion network with reverse attention mechanism

Crowd Density Estimation Using Fusion of Multi-Layer Features

A multi-feature spatial–temporal fusion network for traffic flow prediction

Traffic Flow Prediction Based on Depthwise Separable Convolution Fusion Network

Urban Traffic Congestion Level Prediction Using a Fusion-Based Graph Convolutional Network

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

A Fast Vehicle Counting and Traffic Volume Estimation Method Based on Convolutional Neural Network

Le-SKT: Lightweight traffic density estimation method based on structured knowledge transfer

Understanding Traffic Density from Large-Scale Web Camera Data

High Resolution Speed Estimation for Large-Scale Freeway Based on Data Fusion Technology

DMGF-Net: An Efficient Dynamic Multi-Graph Fusion Network for Traffic Prediction

MF-CNN: Traffic Flow Prediction Using Convolutional Neural Network and Multi-Features Fusion

DSNet: A vehicle density estimation network based on multi-scale sensing of vehicle density in video images

A Multi-Stream Feature Fusion Approach for Traffic Prediction

Spatial-Temporal Traffic Flow Prediction with Fusion Graph Convolution Network and Enhanced Gated Recurrent Units

Multi-scale fusion public gathering recognition based on residual network

Double multi-scale feature fusion network for crowd counting

A lightweight dense crowd density estimation network for efficient compression models