Pedestrian fatalities are on the rise, with more than 6,000 pedestrians killed each year in the United States. There are several technologies and use cases that can help cities make the roads and highways safer. The Smart Intersection proof of concept (POC) deployed by Spectrum is one such example that demonstrates how cities can use technology for protecting pedestrians.
The Smart Intersection proof of concept uses computer vision, edge and near-edge computing to detect and monitor the pedestrian and vehicle movement at the intersection. The anonymous pedestrian and vehicular traffic data is stored in the cloud for further analysis, planning and design of the components of the intersection, including stop signs, traffic/pedestrian light timing, crosswalks and sidewalks to improve pedestrian safety.
This paper discusses the Smart Intersection architecture, machine-learning (ML) model and computer vision technology. The paper also explains the collection, storage, visualization and analysis of metadata on pedestrian and vehicle movement at the intersection.