Skip to Main Content
Objects in video sequence have attracted much attention due to their role in many applications such as tracking, surveillance, and video compression. In this paper, we propose an efficient and accurate method for detecting moving objects in a video sequence. Motion information is used to detect primitive object candidates, so global motion and optical flow estimation is the first step in this algorithm. The estimation is done with our proposed method which is a modification of Horn-Schunck method. The proposed method for optical flow estimation is very fast and needs less iterations than does the original one (Horn-Schunk). Those areas which have different motion from global motion, according to efficient criterion, are marked as object candidates. The spatial information is used to extract the final object. The experimental results show the efficiency and accuracy of proposed algorithm.