Automated rockmass discontinuity mapping from 3-dimensional surface data

• Published in: Engineering geology 2013-09, Vol.164, p.155-162
• Main Authors: Vöge, Malte, Lato, Matthew J., Diederichs, Mark S.
• Format: Article
• Language: English
• Subjects: Applied sciences; Automated algorithms; Buildings. Public works; Exact sciences and technology; Feature extraction; Geologic structure; Geotechnics; LiDAR; Measurements. Technique of testing; Photogrammetry; PlaneDetect; Soil investigations. Testing
• ISSN: 0013-7952; 1872-6917
• DOI: 10.1016/j.enggeo.2013.07.008

Remote sensing technologies, specifically terrestrial-based static LiDAR and photogrammetry, are transforming from state-of-the-art to state-of-practice tools for engineering geologists. The complexity of available software packages to perform standard geomechanical analyses is slowing the widespread adoption of these technologies within the geotechnical community. The development of automated processing tools for feature extraction and data interpretation is aimed at eliminating the need for complex software and manual analysis. This paper presents the development of the algorithms used in the software program PlaneDetect for the automated identification and mapping of planar discontinuities within a 3-dimensional surface model of a jointed rockmass. The software employs a five stage procedure of: surface smoothing, edge detection and masking, blast damaged detection and masking, discontinuity identification, and discontinuity set clustering. The software outputs a stereonet of discontinuity orientations colored by joint set family, an image of the 3-dimensional model with each mapped discontinuity colored by the set family, and a text file of discontinuity orientations. The results of the geomechanical analyses computed by PlaneDetect in comparison to the manual mapping results are more statistically reliable based on less user bias. The time saving realized through using PlaneDetect for mapping discontinuities is approximately ten times compared to the manual mapping approaches. •Development of a new algorithm for automated fracture mapping•Mathematics of the algorithm is explained in detail.•Example liDAR data processed within 2degrees of manual methods•Algorithms developed are 10× more efficient than manual methods.