Evaluation of a Computer Vision Based Automatic Incident and Congestion Detection System in an Urban Context by M.E. Diaz, R Ferris, V. Cavero, S. Guillen, J.J. Martinez, A. Fuertes.
The objective of the document is to describe an Automatic Incident and Congestions Detection System based on computer vision technology, in urban environment. outlines main technical features system presents first evaluation results obtained a field trial currently carried out protect INVAII) II. within Transport Telematics Programme EC. CCTV cameras are installed street intersections lor surveillance purposes. Based image processing techniques. Video Analyzers or Local Sensor Modules (LSM) connected number selected cameras, obtain traffic status measurements view each one them. data further processed at central level, where calculates full network, spatial temporal relations. displays this information operator triggers alarms if any abnormal situation detected. developed by international consortium. was undertaken scale test Valencia. Spain. pilot consists 8 analyzers four intersections, while equipment Traffic Control Center.