Calibration method for deep hole inner surface measurement system based on circular structured light

Abstract Addressing the issues of complex calibration processes, limited accuracy, and reliance on high-precision calibration objects in the calibration of circular structured light measurement systems for deep hole precision measurement, this study proposes a light plane calibration method based on monocular vision and geometric constraints to achieve high-precision mapping from two-dimensional pixel coordinates to three-dimensional contour point clouds. The study uses a two-step alignment strategy to complete the coaxial alignment of the camera optical axis and the ring gauge axis by combining calibration plate pose feedback with visual alignment technology based on ellipticity criteria. The optical plane equation is constructed based on the geometric constraint relationship between the two-dimensional pixel coordinates and the inner hole of the ring gauge, and the three-dimensional coordinate calculation is converted into a single-variable depth parameter optimization problem. With the help of the optical plane normal vector, an orthogonal basis is constructed to achieve local coordinate conversion and radius solution of the cross-section point cloud in the camera coordinate system. The results of the ring gauge inner diameter measurement experiment show that the maximum error is 0.005 mm, the standard deviation is 0.003 mm, and the average relative error is 0.008%. This method significantly improves the operability and engineering applicability of the system and provides a reliable technical solution for the efficient and precise inspection of deep hole parts.