Object Tracking with Kalman Filter and Discrete Cosine Transform

Object tracking is important in the field of computer vision which is used to track targets in video from frame to frame. Kalman filter is a technique which can calculate the predicted position of the target in the next frame by using the history trajectory information. Knowing the predicted position of the target, image template matching is used to get the measured position around the predicted position. Discrete cosine transform (DCT) has a strong energy compaction property, so it can be used to extract features of the target effectively. In this paper, we use DCT to extract features and this does improve the speed of image template matching. Further more, we present a method of adaptively adjusting the threshold determining whether the target is sheltered or not.. The experimental results show that our method outperforms the traditional Kalman filters.