Evidence of Unacceptable Video Detector Performance for Dilemma Zone Protection

Abstract

The use of video imaging vehicle detection systems (VIVDSs) in Texas has increased dramatically, primarily as a result of safety issues and costs. Nonintrusive detectors are almost always safer to install at intersections than inductive loops because of the greater separation between passing motorists and the field crews installing the detectors. Other factors that have contributed to increased VIVDS usage include the flexibility in adjusting detection zones (e.g., with lane reassignments), the ability to send an image of the traffic stream to a traffic operations center, and the lack of damage to the pavement structure (unlike inductive loops, which require a saw-cutting process that weakens the pavement). Despite these advantages, VIVDSs require additional research in some situations to ensure safe operations. The research objective is to determine how well the current video imaging systems deployed by the Texas Department of Transportation provide dilemma zone protection at high-speed signalized intersections. Preliminary findings from data collected at one of the three planned sites indicate that the detection discrepancies between VIVDSs and in-pavement sensors are significantly different. These discrepancies are not always critical to safety but would increase intersection delay.