A Novel Approach for Detection and Tracking of Vehicles using Kalman Filter

Detecting moving objects in videos is an important task in many computer vision applications, including human interaction, traffic monitoring and Structural Health Monitoring. When having a stationary camera, a basic method to detect the objects of interest is background subtraction. However, precise moving object detection using such a method is an extremely difficult task in a varying environment. This paper introduces a new technique for detecting, tracking as well as counting the vehicles based on Kalman filtering approach. In this method background is updated using Kalman filter to detect, track and count the vehicles. The proposed algorithm is tested with different videos and results clearly show the efficiency of the algorithm.