On-line point cloud data extraction algorithm for spatial scanning measurement of irregular surface in copying manufacture

Abstract

For obtaining high-quality point cloud data of a measured surface in copying manufacture, 3D scanning devices typically have extremely high data sampling rate to produce huge amounts of dense points, which lead to significant computational challenges for subsequent data processing tasks in practical applications. Bottlenecks are created owing to inefficiencies in storing, manipulating, and transferring the massive point data. On-line point cloud data extraction is an effective means to solve the above problems. This paper proposes a novel on-line point cloud data extraction algorithm for spatial scanning measurement of irregular surface. The proposed data extraction framework can handle point sets of arbitrary and varying size, point density, and scanning line shape. Furthermore, it can reduce extremely dense point cloud data in ensuring data accuracy during the real-time scanning measuring process. Additionally, we present a bi-Akima method for connecting spatial point sequence in non-planar cross section. It is designed to deal with point cloud data extraction for any kinds of scanning lines during the real-time measuring process. Finally, a series of simulations and real measuring experiments are conducted, and the results show a strong data extraction performance of our proposed method both in data reduction ratio and accuracy.