Intersection Control Through Video Image Processing: Executive Summary

Abstract

Among the most promising and innovative concepts today for alleviating urban traffic congestion is the use of video imaging for vehicle detection, automatic surveillance, and advanced control strategies. Because of its conceptual appeal, research in this area was initiated in the mid 70's in the United States and abroad. A system for vehicle detection through video imaging was recently developed at the University of Minnesota and is being implemented on the 1-394 and l-35W freeways in Minneapolis, Minnesota for incident detection. The Minnesota system, called AUTOSCOPE (TM), emulates loop detectors, a large number of which can easily be placed within the field of the camera's view through interactive graphics. In recent tests its performance matched or exceeded that of loops in vehicle counting, speed measurements, and extraction of certain measures of effectiveness. Evaluation tests of the AUTOSCOPE (TM) were very encouraging, thus the system was installed at a traffic intersection to demonstrate the effectiveness of this new technology as a replacement for loop detectors.