Recommended CitationRhodes, A., D. M. Bullock, J. R. Sturdevant, and Z. T. Clark. Evaluation of Stop Bar Video Detection Accuracy at Signalized Intersections. Publication FHWA/IN/JTRP-2005/28. Joint Transportation Research Program, Indiana Department of Transportation and Purdue University, West Lafayette, Indiana, 2006. doi: 10.5703/1288284313401.
Many agencies nationwide have adopted video vehicle detection technology as an alternative to inductive loops. While many product evaluations have been performed, the majority of these evaluations have concentrated on freeway applications where speed and volume were the primary evaluation criteria. At an actuated intersection, the metrics of speed and volume do not necessarily represent how well a device will operate as a presence detector. Video detection was evaluated at two signalized intersections in West Lafayette, Indiana and Noblesville, Indiana. A camera on each approach was located at the vendor recommended position, at a height of 40 feet and offset so that the camera was approximately inline with the lane line dividing the left-turn lane and through lane. Two additional cameras were located on each approach at less optimal horizontal and vertical locations. Additionally at the West Lafayette site, a fourth camera was located directly above the stop bar so that it looked down onto the detection zone. Inductive loops were used as baseline data to screen for discrepancies. Each time the loop detectors were not in agreement with a specific video detector, a discrepancy was noted. A digital video recording was later observed to determine whether the video detector or the loop detector was in error. An analysis of the data showed video detection was found to produce statistically significantly more false detections and missed detections than the loop detectors on most phases. Generally, only very small variations in video detector performance were observed at the different camera locations. The video detection was also evaluated in terms of its’ consistency of detector turn-on and turn-off times. The video detection displays an undesirable inconsistency between night periods and day periods where it tends to activate 1 to 3 seconds earlier during the night period due to headlight reflections on the pavement. However, the camera at West Lafayette located directly above the stop bar displayed very consistent behavior between the night and day periods.
traffic engineering, traffic signals, traffic surveillance, traffic management, traffic control devices, loops, intelligent transportation systems, statistical analysis, statistics, video detection, vehicle detector performance, SPR-2869
Joint Transportation Research Program
West Lafayette, IN
Date of this Version