A feature extraction method for deformation analysis of large-scale composite structures based on TLS measurement

How to obtain a three-dimensional (3D) model efficiently and extract the feature information of larger-scale composite structures, such as tunnels, accurately is a significant issue in the field of health monitoring. Therefore, an effective method based on TLS measurement is proposed and developed using surface-based non-destructive technology. In this paper, terrestrial laser scanning (TLS) technology is adopted to investigate the tunnel structure, focusing on the extraction of the characteristic section and central curve, which could be applied in deformation monitoring. Point cloud data from TLS measurement is processed in four steps: section extraction, section projection, calculation of central points and curve approximation. The innovation of this paper lies in the projection and iterative filtering of the ring data and rasterization of the point clouds for vertical and horizontal lines. The random sample consensus (RANSAC) algorithm is implemented to approximate the vertical and horizontal lines. The central curve, approximated from the central points, agrees with the general design model and the accuracy falls within the millimeter range.