Point-cloud acquisition and processing for gear profile evaluation based on line-structured light

作者：Xianglei Zhang, Can Wang, Chunxiao You, Wangda Lu, Xin Jiang, Hongming Zhou · 发表于：Engineering Research Express · 年份：2025 · DOI：10.1088/2631-8695/ae25bc · 研究领域：Optical measurement and interference techniques、Advanced Measurement and Metrology Techniques、Surface Roughness and Optical Measurements

Abstract Line structured light technology offers a rapid and non-contact solution for gear inspection. This study proposes a novel method for gear tooth profile error analysis. By establishing a precise spatial and kinematic mapping between the sensor and the gear, a systematic point cloud acquisition procedure is implemented to capture the gear tooth surface topography. The raw point cloud data undergoes preprocessing through an enhanced voxel grid downsampling algorithm, followed by a hole repair process utilizing a radial basis function-based interpolation scheme. Experimental validation on a dedicated testbed demonstrates a high level of repeatability in the profile error curve, with a maximum deviation of 4 μm compared to contact methods. The consistent measurement trends confirm the high accuracy and industrial applicability of the proposed method.