Robust objects tracking algorithm based on adaptive background updating

Abstract

For solving the problem of false tracking by Continuously Adaptive Mean Shift (CAMSHIFT) algorithm when sharing significant color similarity between object and background and changes of object color, moreover, for avoiding selecting initial target object by hand, an adaptive robust objects tracking algorithm based on active camera is proposed. It uses the disparity of global and local motion to detect the motion area. Then, it segments each object by an improved K-Mean clustering algorithm. Finally, it tracks the object by the improved adaptive background updating CAMSHIFT algorithm continuously in real time. The effectiveness of this proposed algorithm has been proved by preceding experiments on real time video sources. Compared to the state of the art methods, the algorithm in this paper is more robust and effective.