An algorithm for automatic detection and orientation estimation of planar structures in LiDAR-scanned outcrops

The spatial orientation of linear and planar structures in geological fieldwork is still obtained using simple hand-held instruments such as a compass and clinometer. Despite their ease of use, the amount of data obtained in this way is normally smaller than would be considered as representative of the area available for sampling. LiDAR-based remote sensors are capable of sampling large areas and providing huge sets of digitized spatial points. However, the visual identification of planes in sets of points on geological outcrops is a difficult and time-consuming task. An automatic method for detecting and estimating the orientation of planar structures has been developed to reduce analysis and processing times, and to fit the best plane for each surface represented by a set of points and thus to increase the sampled area. The algorithm detects clusters of points that are part of the same plane based on the principal component analysis (PCA) technique. When applied to real cases, it has shown high precision in both the detection and orientation of fractures planes.