Performance comparison of object tracking methods on video frames using parallel computing

Moving object detection used in wide range of application especially in automated traffic surveillance and management. In defense, these techniques are used in automating threat detection and elimination, target acquisition and also in underwater object tracking. Moving object detection is carried using various methods and an analysis is made to know its precision of detection. The methods on which it is implemented is Kernel Density Estimation, Gaussian Mixture Model, Visual Background Extraction, Multi cue Background Subtraction. The experiment is carried out on different videos and Vibe is found to give better result.