An unsupervised manifold learning framework for automated identification of rock discontinuities in underground mines using raw LiDAR point cloud
|
Authors: Han, Z; Schwartzkopff, A; Mathew, TJ; Moridi, M; Akdag, S |
DOI https://doi.org/10.36487/ACG_repo/2625_27
Cite As:
Han, Z, Schwartzkopff, A, Mathew, TJ, Moridi, M & Akdag, S 2026, 'An unsupervised manifold learning framework for automated identification of rock discontinuities in underground mines using raw LiDAR point cloud', in D Cumming-Potvin & J Wesseloo (eds), Ground Support 2026: Proceedings of the 11th International Conference on Ground Support in Mining, Australian Centre for Geomechanics, Perth, pp. 1-11, https://doi.org/10.36487/ACG_repo/2625_27
Abstract:
Accurate mapping of rock discontinuities is essential for rock mass characterisation, stability analysis and ground-support design. Traditional manual approaches are time-consuming, labour-intensive, and subjective in interpretation. Digital approaches based on photogrammetry and Light Detection and Ranging (LiDAR) point cloud methods provide efficient tools for solving the problem. However, photogrammetry-based methods are challenging due to the low-light and dusty underground environments, making LiDAR point cloud data a superior tool to the traditional approach, which can actively emit lasers and is therefore independent to the illumination conditions. Supervised deep learning-based methods are widely adopted for 3D point cloud analysis in underground constructions due to their robust performance. Although the performance of these models is enhanced by adopting advanced deep learning strategies, the application and generalisation ability of these methods are limited by the supervised manner, which requires extensive labelled datasets. Therefore, this study proposes a novel unsupervised deep neural network-based method (ManiRock) for automatic rock discontinuities identification from LiDAR point clouds. We integrate a state-of-the-art unsupervised implicit learning method and a widely used clustering method. ManiRock directly takes the raw point cloud as input and constructs a continuous implicit representation of the rock surface manifold and identify and isolate each discontinuity subsequently in a two-step manner. This study provides a case study as a geometric proofofconcept using limited unsupported underground tunnel rock mass point clouds. The method is tested using LiDAR point cloud datasets acquired from a metal mine located in Kalgoorlie, Western Australia. The results are compared with ground-truth results derived from manual mapping and demonstrate that ManiRock achieves high-fidelity discontinuity mapping while bypassing the need for data labelling, providing a potentially useful tool for rapid geotechnical assessment of underground rock mass.
Keywords: rock mass discontinuities, LiDAR, unsupervised manifold learning
References:
Assali, P, Grussenmeyer, P, Villemin, T, Pollet, N & Viguier, F 2016, ‘Solid images for geostructural mapping and key block modeling of rock discontinuities’, Computers & Geosciences, vol. 89, pp. 21–31,
Battulwar, R, Emami, E, Zare, M, Battulwar, K, Shahsavar, M, Moniri-Morad, A & Sattarvand, J 2024, ‘Utilizing deep learning for the automated extraction of rock mass features from point clouds’, Geotechnical and Geological Engineering, vol. 42, no. 7, pp. 6179–6194,
Battulwar, R, Zare-Naghadehi, M, Emami, E & Sattarvand, J 2021, ‘A state-of-the-art review of automated extraction of rock mass discontinuity characteristics using three-dimensional surface models’, Journal of Rock Mechanics and Geotechnical Engineering, vol. 13, no. 4, pp. 920–936,
Boon, CW, Houlsby, GT & Utili, S 2014, ‘Designing tunnel support in jointed rock masses via the DEM’, Rock Mechanics and Rock Engineering, vol. 48, no. 2, pp. 603–632,
Chen, J, Fang, Q, Zhang, D & Huang, H 2023, ‘A critical review of automated extraction of rock mass parameters using 3D point cloud data’, Intelligent Transportation Infrastructure, vol. 2,
Chen, J, Yang, T, Zhang, D, Huang, H & Tian, Y 2021, ‘Deep learning based classification of rock structure of tunnel face’, Geoscience Frontiers, vol. 12, no. 1, pp. 395–404,
Erharter, GH, Bar, N, Hansen, TF, Jain, S & Marcher, T 2024, ‘International distribution and development of rock mass classification: A review’, Rock Mechanics and Rock Engineering, vol. 58, no. 10, pp. 11169–11180,
Ester, M, Kriegel, H-P, Sander, J & Xu, X 1996, ‘A density-based algorithm for discovering clusters in large spatial databases with noise’, Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, AAAI Press, Menlo Park, pp. 226–231.
Fecker, E & Rengers, N 1971, ‘Measurement of large scale roughnesses of rock planes by means of profilograph and geological compass’, in R Houpert (ed.), Rock Fracture: Proceedings of the International Symposium on Rock Mechanics, Ecole Nationale Superieure de Geologie appliquee et de Prospection miniere, Vandœuvre-lès-Nan.
Ferrero, AM, Migliazza, MR & Umili, G 2014, ‘Rock mass characterization by means of advanced survey methods’, ISRM-EUROCK-2014-002, in ISRM Regional Symposium – EUROCK 2014, Vigo.
Ge, Y, Wang, H, Liu, G, Chen, Q & Tang, H 2025, ‘Automated identification of rock discontinuities from 3D point clouds using a convolutional neural network’, Rock Mechanics and Rock Engineering, vol. 58, no. 3, pp. 3683–3700,
Gigli, G & Casagli, N 2011, ‘Semi-automatic extraction of rock mass structural data from high resolution LIDAR point clouds’, International Journal of Rock Mechanics and Mining Sciences, vol. 48, no. 2, pp. 187–198,
Gomes, RK, de Oliveira, LPL, Gonzaga, L, Tognoli, FMW, Veronez, MR & de Souza, MK 2016, ‘An algorithm for automatic detection and orientation estimation of planar structures in LiDAR-scanned outcrops’, Computers & Geosciences, vol. 90, pp. 170–178,
Han, Z, Li, Q, Wang, X & Li, D 2025, ‘A physics-guided hierarchical deep learning framework for underground rock reinforcement compliance check based on 4D point cloud data’, Tunnelling and Underground Space Technology, vol. 164,
Han, Z, Lin, W, Li, Q & Li, D 2026, ‘Full-field deformation quantification of underground tunnels using SLAM LiDAR point cloud based on unsupervised neural implicit learning’, Tunnelling and Underground Space Technology, vol. 175,
Kinde, M, Getahun, E & Jothimani, M 2024, ‘Geotechnical and slope stability analysis in the landslide-prone area: a case study in Sawla – Laska road sector, Southern Ethiopia’, Scientific African, vol. 23,
Lague, D, Brodu, N & Leroux, J 2013, ‘Accurate 3D comparison of complex topography with terrestrial laser scanner: application to the Rangitikei canyon (N-Z)’, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 82, pp. 10–26,
Lato, MJ & Vöge, M 2012, ‘Automated mapping of rock discontinuities in 3D lidar and photogrammetry models’, International Journal of Rock Mechanics and Mining Sciences, vol. 54, pp. 150–158,
Li, C, Zhou, J, Du, K & Tao, M 2025a, ‘Enhanced discontinuity characterization in hard rock pillars using point cloud completion and DBSCAN clustering’, International Journal of Rock Mechanics and Mining Sciences, vol. 186,
Li, K, Zhao, M, Wu, H, Yan, D-M, Shen, Z, Wang, F-Y & Xiong, G 2022, ‘GraphFit: Learning multi-scale graph-convolutional representation for point cloud normal estimation’, The Proceedings of the 17th European Conference on Computer Vision, Springer, Cham, pp. 651–667,
Li, Q, Feng, H, Gong, X & Liu, Y-S 2025b, ‘Learning normals of noisy points by local gradient-aware surface filtering’, 2025 IEEE/CVF International Conference on Computer Vision (ICCV), The Institute of Electrical and Electronics Engineers, Danvers,
Li, Q, Feng, H, Shi, K, Gao, Y, Fang, Y, Liu, Y-S & Han, Z 2024, ‘Learning signed hyper surfaces for oriented point cloud normal estimation’, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 46, no. 12, pp. 9957–9974,
Li, X, Yao, M, Yuan, J-d, Wang, Y-j & Li, P-y 2023, ‘Deep learning characterization of rock conditions based on tunnel boring machine data’, Underground Space, vol. 12, pp. 89–101,
Maria, FA, Maria, M & Paolo, GG 2004, ‘Analysis of tunnel stability: comparison between continuous and discontinuous approaches’, International Journal of Rock Mechanics and Mining Sciences, vol. 41, pp. 646–651,
Milne, D, Hadjigeorgiou, J & Pakalnis, R 1998, ‘Rock mass characterization for underground hard rock mines’, Tunnelling and Underground Space Technology, vol. 13, no. 4, pp. 383–391,
Palmström, A 1996, ‘Characterizing rock masses by the RMi for use in practical rock engineering: Part 1: The development of the Rock Mass index (RMi)’, Tunnelling and Underground Space Technology, vol. 11, no. 2, pp. 175–188.
Priest, SD 1993, Discontinuity analysis for rock engineering, Springer, Dordrecht.
Pu, C, Zhan, J, Huang, D, Zhu, W & Peng, J 2025, ‘Interpretation and characterization of sedimentary rock mass structure using UAV multi-sensor data’, Rock Mechanics and Rock Engineering, vol. 58, no. 7, pp. 8415–8443,
Singh, SK, Raval, S & Banerjee, BP 2021, ‘Automated structural discontinuity mapping in a rock face occluded by vegetation using mobile laser scanning’, Engineering Geology, vol. 285,
Tang, N, Wang, L, Jiang, H, Huang, X, Tan, G & Zhou, X 2023, ‘A new clustering method of rock discontinuity sets based on modified K-means algorithm’, Bulletin of Engineering Geology and the Environment, vol. 82, no. 11, pp. 415,
Wang, R, Zhan, J, An, X, Yao, Z & Peng, J 2026, ‘Efficient method for identification of rock mass discontinuities using 3D point cloud data’, Rock Mechanics and Rock Engineering, in press,
Xu, H & Tang, S 2025, ‘Digital tunnel geometry model (DTGM): A multimodal data fusion framework for rock mass feature quantification’, International Journal of Rock Mechanics and Mining Sciences, vol. 194,
Yi, X, Feng, W, Wang, D, Yang, R, Hu, Y & Zhou, Y 2023, ‘An efficient method for extracting and clustering rock mass discontinuities from 3D point clouds’, Acta Geotechnica, vol. 18, no. 7, pp. 3485–3503,
Zhan, J, Chen, J, Xu, P, Zhang, W, Han, X & Zhou, X 2017a, ‘Automatic identification of rock fracture sets using finite mixture models’, Mathematical Geosciences, vol. 49, no. 8, pp. 1021–1056,
Zhan, J, Xu, P, Chen, J, Wang, Q, Zhang, W & Han, X 2017b, ‘Comprehensive characterization and clustering of orientation data: A case study from the Songta dam site, China’, Engineering Geology, vol. 225, pp. 3–18,
Zhou, J-w, Chen, J-l & Li, H-b 2024, ‘An optimized fuzzy K-means clustering method for automated rock discontinuities extraction from point clouds’, International Journal of Rock Mechanics and Mining Sciences, vol. 173,
Zhou, Z, Yu, J, Gao, J, Cai, X & Wang, Z 2025, ‘Rapid rock mass classification framework using machine learning and semantic segmentation for underground mining’, Engineering Applications of Artificial Intelligence, vol. 159,
© Copyright 2026, Australian Centre for Geomechanics (ACG), The University of Western Australia. All rights reserved.
View copyright/legal information
Please direct any queries or error reports to repository-acg@uwa.edu.au
View copyright/legal information
Please direct any queries or error reports to repository-acg@uwa.edu.au