Evaluation of LIDAR systems for rock mass discontinuity identification in underground stone mines from 3D point cloud data

摘要

The goal of this thesis is to compare the accuracy and precision of the discontinuity identification results obtained by three active remote sensing technologies along with a point cloud processing program, and the results obtained by conventional methods. For this research, the active remote sensing devices were terrestrial LIDAR, mobile LiDAR with Simultaneous localization, and mapping (SLAM), and LIDAR/Camera on an autonomous UAV. The open-source point cloud data processing programs Discontinuity Set Extractor (DSE) and the Cloud Compare were used to process point cloud data. The results of this research found that it is possible to identify certain geological structures such as bedding planes with even a point cloud density of 0.5 points per cm square, from the point intensity. However, identifying joint sets require higher point densities and needs detailed analysis of the 3D maps and expert interpretation. Therefore, this thesis couldn’t conclude if only LIDAR measurements, without expert interpretation, would be enough to identify geological structures even with high point densities. However, point intensity together with the high point density will allow more accurate identification of the geological structures. Hence, LIDAR camera used on the autonomous robotic system can provide both accurate point coordinates with LIDAR measurement, but it requires necessary illumination to obtain clear pictures with the camera to perform an appropriate identification of the geological structures from dense 3D maps possible.