The development of a machine vision system for adaptive bending of sheet metals

Abstract Press-brake bending is a common forming process widely used in the sheet metal industry. One of the major problems in the bending operation is the control of the amount of springback. A sheet metal bending machine which is capable of on-line measurement of the punch load and punch displacement data has been built. A computer vision system is incorporated in the machine to capture the image of the bend, carry out features extraction and image analysis to find out the profile of the bend. The image data are fed into a plastomechanical model which will compute the required punch travel to achieve a given angle of the bend. These image data can also be used to monitor the bending process and control the product quality. The inherent variation in material properties is taken into account during each bending operation.