The Harsh Truth: Segment-Level Analysis of Harsh Driving Events in Milan Using Large-Scale Telematics, Street Networks, and Google Street View
Researchers have developed a new method to analyze harsh driving events in Milan by combining telematics data, street network information, and Google Street View imagery. This approach uses semantic segmentation and machine learning to identify factors associated with increased harsh acceleration and braking. The study found that wider roads, intersections, and open visual fields correlate with higher harshness, while denser built environments are linked to lower harshness. The findings suggest that targeted safety interventions, rather than uniform approaches, are needed for urban road safety. AI
IMPACT This research demonstrates how AI-powered image analysis can enhance urban planning and road safety interventions.