A Machine Learning Approach to Extract Rock Mass Discontinuity Orientation and Spacing, from Laser Scanner Point Clouds

This study wants to give a contribution the semi-automatic evaluation of rock mass discontinuities, orientation and spacing, as important parameters used in Engineering. In complex inaccessible areas, traditional geological survey is hard conduct, therefore, remote sensing techniques have proven be very useful tool for discontinuity analysis. However, critical expert judgment necessary make reliable analyses. For this reason, open-source Python named DCS (Discontinuities Classification Spacing) was developed manage point cloud data. The written based on semi-supervised clustering. By approach users can: (a) estimate number sets (here referred “clusters”) using Error Sum Squares (SSE) method K-means algorithm; (b) evaluate step by quality classification visualizing stereonet scatterplot dip vs. direction from clustering; (c) supervise clustering procedure through manual initialization centroids; (d) calculate normal spacing. contrast other algorithms available literature, does not require inputs permits at each step. tested steep coastal cliff Ancona town (Italy), called Cardeto–Passetto cliff, which characterized fracturing largely affected rockfall phenomena. results were validated with field compared ones FACETS plug-in CloudCompare. addition, algorithm regular surfaces an anthropic wall located bottom cliff. Eventually, kinematic analysis slope stability performed, discussing advantages limitations methods considered making fundamental considerations their use.