Tabu search based implementation of object tracking using joint color texture histogram

Abstract

Object tracking has many applications like security and surveillance, traffic control and many others. In this Paper new methodology has been proposed which uses tabu search algorithm along with joint color texture histogram to track object. Texture featurelike edges and corners are extracted using uniform linear binary pattern, which will help in increasing the robustness of the object to be tracked. After features are extracted battacharyya coefficient is calculated to measure similarity between target's raw feature in the previous frame and raw feature in present frame. This coefficient is used as objective function in tabu search algorithm framework. In tabu search, parameters like length of tabu list and radius of search region are chosen appropriately. The experimental results shows that the proposed algorithm can track object more robustly under complex scenario's like object is similar to background in a static video.