Video-Based Detection and Tracking Model for Traffic Surveillance
Abdul Razak Al Kherret, Al Sayed Al Sobky, Ragab M. Mousa · Transportation Research Board 94th Annual MeetingTransportation Research Board · 2015
The continuous increase of traffic congestion in urban areas demands a high reliable traffic management system for monitoring traffic flows and providing key input parameters for predicting traffic conditions. Video sequences of road scenes are increasingly used in several contexts with an emphasis on automation, notably for tracking moving objects in a static background. This paper presents a multiple-vehicle surveillance model developed using Matlab Software for detecting and tracking moving vehicles, and collecting traffic data for different lengths of region of interest (ROI), ranging between 5 and 30 m. The model was validated using simulated video scenes, designed in VISSIM with known traffic data. Measurements from model were compared with actual measurements reported by VISSIM and results confirmed exact match of vehicle counts. Statistical t-tests of mean speed differences confirmed the model validity at 5% significance level, especially with ROI length of 10 and 15 m. Validation of headway measurements was also confirmed for optimum ROI lengths. The model processes one second in video clips of frame rate 20 frames/sec in 0.96 sec. This is appropriate for real-time applications to yield traffic parameters including vehicle speed, headway, count, incident detection, queue detection, etc. However, the model was validated assuming no lane changes and no overlap of vehicles, and, hence, model validity is limited to these assumptions. It is recommended that this model be validated using real world videos containing noises such as light variation, shadows, vibrations due to wind, skewed views, lane changes, and/or trucks that obscure full view of vehicles.