A New Color Based Optical Flow Algorithm for Environment Mapping Using a Mobile Robot

Abstract

Environment mapping from a video sequence is considered to be one of the most important problems in computer vision because of its application in surveillance, virtual reality, autonomous navigation, multimedia communications, medical prognosis, etc. In this paper, we have presented an optical flow based method for environment mapping. It uses a new color based optical flow computation technique. The camera, which is mounted on a mobile robot, is kept perpendicular to the direction of motion, and the captured set of images is used to compute the dense depth map. We have used a Kalman filter to denoise the depth map.