Deep Hole Inner Surface Three-dimensional (3-D) Reconstruction by a Multi-line Structured Laser Light Projection System
S. Pan, X-J. Shao, L-L. Yang, B. Song and X-J. Tang
The three-dimensional (3-D) reconstruction of the inner surface of the deep hole is of great significance for the quality inspection, fault diagnosis and life prediction of deep hole parts. Using semiconductor laser light, the development of a novel 3-D reconstruction system to generate the 3-D point cloud of the inner surface of the deep hole is described herein. The measurement is carried out in the narrow space inside the deep hole, and the surface point clouds of different positions are obtained and spliced into a unified coordinate frame. In a single perspective, dozens of laser light planes are projected on the deep hole inner surface, the camera captures the image of the laser light stripes on the inner surface, the 3-D point cloud of the inner surface is solved by the parameters of the camera and the projected light planes. The point cloud acquirement system is driven by the stepping motor to scan the entire surface inside the deep hole, a coordinate transformation model is proposed to provide a rough position of each single-view point cloud, and the point cloud registration algorithm is used to optimize the reconstruction result. A calibration experiment is carried out to obtain the system parameters required for 3-D reconstruction, and the measurement experiments indicate that the rotation reconstruction method is applicable in inner surface measurement and inspection.
Keywords: Semiconductor laser, deep hole, multi-line, three-dimensional (3-D) reconstruction, iterative closest point (ICP) algorithm, point cloud