Publication

Abstract : Object tracking in video sequencesis gaining much importance in computer vision. Many video surveillance system implemented template or pattern matching to track the moving object.However, template or pattern matching requires sample of pattern or template before it can track any object. Object tracking can be difficult if the surface of the object does not remain same from frame to frame, such as reflection of light. This paper proposes a new method to detect a single moving object in a multimedia file such as video through an approach called Optical flow which proves to be a more efficient method for tracking moving objects.Optical flow is a flexible representation of visual motion that is particularly suitable for computers for analyzing digital images. In this work the Horn & Schunck method is used to find the optical flow vectors which in turn pave a way for the detection and tracking of the single moving object in a video. Optical flow method introduced in this work proves to be more efficient method for object detection and tracking compared to existing methods of object detection tracking. A detailed study on the existing methods is presented. The proposed method was experimented with various videos and the results are also discussed.