With continually increasing development of the industrial piece manufacturing technology the necessity of using high precision measurement devices is appeared. Machine vision based methods are a group of new approaches for measuring industrial pieces. These methods function by capturing image from the piece, so they don’t need to be in contact with the piece during the measurement process and they have acceptable precision. When the total dimensions of the piece can’t be fitted in one image frame, it is necessary to capture several images from different regions of the piece in order to cover all areas of the piece. In this paper a new precise method for connecting captured images is proposed. In the proposed method two adjacent images are connected based on the recognizing their common part
