Linear Structured Light System with Optimized Laser Stripe Extraction for Measurement of Large-Sized Workpieces

Abstract

At present, the commonly used three-dimensional shape measurement technology of large-sized workpieces has the problems of poor robustness, low measurement efficiency and high cost. In order to solve the above problems, we propose a new measurement method which combines linear structured light and stereoscopic tracking. This paper mainly describes the linear structured light measurement system. A secondary extraction algorithm is designed based on the normal direction according to the commonly used laser stripe center line extraction algorithm. The algorithm improves the Gaussian curve fitting method. By combining the fine state refinement method, Hessian matrix method, Gaussian curve fitting and light plane calibration, the fast extraction of the center line of the laser stripe is completed, and the mapping relationship between the two-dimensional coordinates of the points on the laser center line and the three-dimensional coordinates in the camera coordinate system is obtained. We conducted feasibility experiments, and the preliminary experimental results indicate that this approach is valid, with measurement errors controlled within 0.44%.