Vehicle Speed Detection System using Motion Vector Interpolation

Abstract

Vehicle speed detection system is used to estimate the speed of a moving vehicle at a given instant of time. The current technology using radar gun is more expensive and it can detect only one object in an instance, with lesser accuracy. Hence, an alternative vehicle speed detection system based on video frame extraction and motion detection is proposed. The moving vehicles are detected and tracked using Motion Vector Interpolation technique. The speed is calculated and displayed for each vehicle in the input video which is for every 0.6 seconds. The simulation is performed using MATLAB and system proved to be an efficient one producing a result with minimum error of less than 3%.