Intelligent CCTV for Urban Design: AI-Based Analysis of Soft Infrastructure at Intersections
Existing security cameras + deep learning = evidence-based urban design at low cost.
An AI analytics framework repurposes existing CCTV feeds to evaluate soft traffic interventions like pedestrian refuges and curb extensions. Using deep learning and perspective-based speed estimation, the study measured driver behavior before and after interventions, with repeated post-installation monitoring in Week 1 and Week 2, in Minneapolis. At unsignalized intersections, mean and 85th-percentile speeds fell by up to 18.75% and 16.56%, while pass-through traffic decreased by as much as 12.2%. Signalized intersections showed comparable reductions except one location, with mean and 85th-percentile speeds dropping by up to 20.0% and 17.19%. These results demonstrate the traffic-calming effectiveness of soft infrastructure and the utility of AI-powered methods for rapid, low-cost transport policy evaluation.
- AI analysis of existing CCTV at 11 Minneapolis intersections showed mean speed reductions up to 18.75% at unsignalized and 20% at signalized intersections after soft infrastructure installation.
- Pass-through traffic volume decreased by 12.2% at unsignalized crossings, indicating fewer drivers cutting through residential areas.
- The framework uses perspective-based speed estimation from regular security cameras, enabling low-cost, repeatable evaluations without installing new sensors.
Why It Matters
AI can turn ordinary security cameras into traffic-calming sensors, enabling cities to test interventions cheaply and at scale.