A Tracking-Based Traffic Performance Measurement System for Roundabouts and Intersections

13744952012Final ReportPDFTech ReportCTS 12-10CTS Project #2011013Intelligent transportation systemsCamerasTraffic surveillanceAutomatic data collection systemsTraffic dataImplementationAlgorithmsVehicle trajectoriesRoundaboutsHighwaysAutomated vehiclesI72: Traffic and Transport PlanningIntersectionsTraffic patternsUnited StatesMinnesotaUniversity of Minnesota. Intelligent Transportation Systems InstituteTang, HuaDinh, HaiUniversity of Minnesota, Duluth. Northland Advanced Transportation Systems Research LaboratoriesUniversity of Minnesota, DuluthBlackketter, D.US Transportation CollectionAutomatic traffic data collection can significantly save labor work and cost compared to manual data collection. The collected traffic data are necessary for traffic simulation and modeling, performance evaluation of the traffic scene, and eventually (re)design of the traffic scene. However, automatic traffic data collection has been one of the challenges in Intelligent Transportation Systems (ITS). This project presents the development of a single camera-based video system for automatic traffic data collection for roundabouts and intersections. The system targets roundabouts and intersections because no mature data collection systems exist for these traffic scenes yet in contrast to highway scenes. The developed system has mainly processing modules. First, the camera is calibrated for the traffic scene of interest and a novel circle-based calibration algorithm is proposed for roundabouts. Second, the system tracks vehicles from the video by incorporating powerful imaging processing techniques and tracking algorithms. Finally, the resulting vehicle trajectories from vehicle tracking are analyzed to extract the interested traffic data, which includes vehicle volume, vehicle speed (including acceleration/de-acceleration behavior), travel time, rejected gaps, accepted gaps, follow-up time and lane use. Practical tests of the developed system show that it can reliably track vehicles and provide reasonably accurate traffic data in most cases