Object Tracking Using Computer Vision Techniques

Abstract

  Computer vision basically deals with different factors such as modeling of camera, lighting, color, texture, shape and motion that affect images and videos from visual inputs. Object detection and tracking are important components in many computer vision applications including activity recognition, traffic monitoring and automotive safety. This paper is about locating a moving object (or multiple objects) over a time using a stationary camera and associating the target object detections in consecutive video frames. In this perspective a video is captured by digital camera and is used for motion analysis. In the first stage of experiment frame differencing algorithm is chosen for object detection and its motion is estimated by associating the centroid of the moving object in each differenced frame. In the second stage of experiment same algorithm is chosen for object detection but motion of each, track is estimated by Kalman filter. However the best estimate is made by combining the knowledge of prediction and correction mechanisms that were incorporated as part of Kalman filter design. The tracking results obtained from two stages of experiment are presented for discussion.