Real-time Measurement of Sheet Metal Bending Angle Based on Monocular Vision

_ Aiming at the problem of online and real-time measurement of sheet metal bending angle, a monocular vision measurement method is proposed in this paper. The principle of this method is that a light beam is projected onto a plane to form a light stripe. When the relative angle between the reflecting plane and the light beam changes, the inclination angle φ of the light stripe changes synchronously. According to this characteristic, this paper proposes to monitor the change of the light stripe through a camera, reverse study the change of the inclination angle θ of the reflecting plane, and achieve online measurement of the bending angle Φ. In this paper, the geometric relationship between the inclination angle φ of the light stripe and the bending angle Φ is derived, and a bending angle measurement model based on two-dimensional light stripe information is established. An optimized gray centroid algorithm is used to identify the light stripe centerline, and a subpixel-level light stripe center extraction algorithm is established. Intelligent algorithms are used for parameter correction to reduce the impact of calibration errors. The bending angle measurement model test was completed. The analysis results show that the method proposed in this paper is correct and feasible, and can achieve high-precision, low-cost, and rapid bending angle measurement. After further improvement, the method proposed in this paper is suitable for online measurement of bending angle and can provide a methodological basis and reference for the development of bending CNC systems. Keywords sheet metal bending, bending angle measurement, geometric model, measurement model, light stripe recognition