Publications

Abstract : Coupling computer vision techniques with automated image acquisition systems is an emerging field of research in recent years. Integration of multiple vision sensors made considerable improvements in image processing and analysis techniques. An attempt is made in this work to establish a relationship between camera motions and the corresponding image acquired. Camera motion is obtained by fixing the same with an existing ABB-IRB 120 manipulator. Motion of the camera is measured by the optical flow field of pixels in image plane. The optical flow field depicts the velocity of pixel in an image with respect to the manipulator motion. Pixel velocity is a function of depth, point coordinates and camera parameters. The methodology employs a camera calibration app for calibrating the camera parameters with an automatic procedure to formulate the image space variables with joint space variables. Based upon the derived relation, sampling of images towards camera calibration is linked with the robot manipulator for every motion and the method allows solving accurately gives ample geometric and algebraic sights.