Analyzing road traffic crashes through multidisciplinary video data approaches

Haiqing Zhang ¹, Yongqiang Shang ^2,3

⁰College of Information Engineering, Shaanxi A&F Technology University, Yangling, Shaanxi, China.

Frontiers in Public Health +

|

December 29, 2025

View abstract on PubMed

Summary

This study introduces a new AI framework using video analytics to understand complex traffic interactions and improve road safety. The data-driven approach identifies risk factors and behaviors preceding accidents in urban environments.

Area Of Science

Urban planning and traffic safety analysis.
Artificial intelligence and machine learning applications.
Computer vision and spatiotemporal data modeling.

Background

Road traffic crashes are a growing global concern, impacting public health and urban safety.
Complex interactions in dynamic urban traffic systems are difficult to model with traditional methods.
There is a need for data-driven frameworks to analyze intricate real-world traffic scenarios.

Purpose Of The Study

To propose a novel computational framework for enhanced traffic safety analysis.
To integrate video data analytics with artificial intelligence for a multidisciplinary approach.
To address the challenges of complex, heterogeneous traffic systems.

Main Methods

Utilizing advanced video data analytics and artificial intelligence techniques.
Employing spatiotemporal modeling, behavioral analysis, and environmental context.
Leveraging large-scale, in-situ video data from urban intersections and road networks.

Main Results

Providing a granular understanding of risk factors and interaction patterns.
Identifying precursors to collisions and near-miss events.
Demonstrating a data-driven approach to traffic safety.

Conclusions

The developed framework enhances the analysis of traffic safety in complex urban environments.
It supports risk-sensitive, behavior-aware decision-making in urban mobility.
The approach is tailored for heterogeneous traffic systems and AI-driven insights.

Keywords:

multi-agent systems neural relational networks risk modeling traffic safety video analytics