Key Takeaways & Executive Findings
- •• SAM achieves high accuracy in 2D discontinuity segmentation with 0.78 mean IoU and 0.86 average precision on a large dataset. • The proposed SAM-SfM framework extends identification to 3D, achieving 0.91 average precision on benchmark datasets. • The framework leverages the inherent pixel-point cloud relationship in SfM, simplifying and generalizing across photogrammetric devices. • The method provides a practical, efficient solution for rock mass discontinuity identification, valuable for geological investigations and data annotation.
Abstract
The identification of rock mass discontinuities is critical for rock mass characterization. While high-resolution digital outcrop models (DOMs) are widely used, current digital methods struggle to generalize across diverse geological settings. Large-scale models (LSMs), with vast parameter spaces and extensive training datasets, excel in solving complex visual problems. This study explores the potential of using one such LSM, Segment anything model (SAM), to identify facet-type discontinuities across several outcrops via interactive prompting. The findings demonstrate that SAM effectively segments two-dimensional (2D) discontinuities, with its generalization capability validated on a dataset of 2426 identified discontinuities across 170 outcrops. The model achieves 0.78 mean IoU and 0.86 average precision using 11-point prompts. To extend to three dimensions (3D), a framework integrating SAM with Structure-from-Motion (SfM) was proposed. By utilizing the inherent but often overlooked relationship between image pixels and point clouds in SfM, the identification process was simplified and generalized across photogrammetric devices. Benchmark studies showed that the framework achieved 0.91 average precision, identifying 87 discontinuities in Dataset-3D. The results confirm its high precision and efficiency, making it a valuable tool for data annotation. The proposed method offers a practical solution for geological investigations.
1. Introduction
A rock mass is a complex fractured system and consists essentially of two constituents: discontinuities, i.e., fractures, fissures, joints, faults, bedding planes and micro-fissures, and intact rock [1]. Discontinuities play a significant role in the mechanical properties of rock masses, since they define the weak planes in a rock mass along which the rock blocks detach and fail [2]. Therefore, the characterization of discontinuities in rock exposures is an important step required to collect input information for further rock mechanics analysis and rock engineering design.
To characterize discontinuities visible at the surface of rock outcrops, a typical set of parameters often suggested for measurement includes orientation, spacing, persistence, and roughness [3–5]. This decisive information on rock discontinuities helps to predict the flow of groundwater, assesses rock mass quality, and provides a better stability condition for rock engineering [6,7]. Traditionally, these parameters are measured manually on-site at rock exposures, for example, measuring discontinuity orientation with a compass and an inclinometer [8]. However, such methods involve trade-offs between the number of samples and labor efficiency, and are often inaccessible and dangerous. To overcome these limitations, three-dimensional (3D) digital outcrop models (DOMs) [9] offer a high-resolution digital twin of rock outcrops.
Loading authentic research manuscript (Pages 1–5)...
Jiawei Wang, Jun Zheng, Jie Hu, Xiaojin Gong, Qing Lü, Ju Han, Jialiang Sun (2025). An interactive framework integrating segment anything model and structure-from-motion for three-dimensional discontinuity identification in rock masses. SinoTechIntel Verified Research. https://doi.org/10.1016/j.ijmst.2025.09.005
Research & Educational Purpose Only:The translations, structured abstracts, analytical annotations, and data reports provided by SinoTechIntel are intended exclusively for academic research, internal corporate R&D, and educational benchmarking. They do not constitute formal engineering, chemical safety, legal, or professional advice.
Copyright & Intellectual Property Notice: Original copyright of the underlying source articles and experimental data remains with the respective authors, institutions, and original publishing journals. SinoTechIntel claims intellectual property only over its proprietary translations, analytical syntheses, and AEO structured enhancements in accordance with international fair use and academic citation principles.
Frequently Asked Questions
What is the main contribution of this paper?
The paper proposes an interactive framework integrating the Segment Anything Model (SAM) with Structure-from-Motion (SfM) for three-dimensional identification of rock mass discontinuities, achieving high precision and efficiency.
How does the framework work?
The framework uses SAM for 2D segmentation of discontinuities in images and leverages the inherent relationship between image pixels and point clouds in SfM to extend the identification to 3D, simplifying the process and generalizing across photogrammetric devices.
What are the performance metrics?
In 2D, SAM achieves 0.78 mean IoU and 0.86 average precision. In 3D, the framework achieves 0.91 average precision, identifying 87 discontinuities in the benchmark dataset.
What is the significance of this method?
The method provides a practical solution for geological investigations, offering high precision and efficiency for rock mass discontinuity identification, and is valuable for data annotation.
What are the potential applications?
The method can be applied in rock mechanics analysis, rock engineering design, groundwater flow prediction, and rock mass quality assessment, among other geological and engineering applications.
Related Technical Papers & Translations
A Novel Approach for Enhanced Brain Tumor Segmentation Using Multimodal MRI and Deep Learning
Brain tumor segmentation from multimodal MRI is crucial for diagnosis and treatment planning. In this study, we propose a novel deep learning framework that integrates structural and functional imaging modalities to improve segmentation accuracy. Our method employs a multi-scale attention mechanism and a hybrid loss function to handle class imbalance and boundary ambiguity. Evaluated on the BraTS benchmark, our approach achieves state-of-the-art performance, with Dice scores of 0.91, 0.87, and 0.84 for whole tumor, core, and enhancing tumor, respectively. Furthermore, we demonstrate the generalizability of our model across different scanners and protocols. Our findings suggest that the proposed method can significantly aid clinical decision-making and surgical planning.
Investigation of coupled acoustic and electrical responses and early warning approaches during re-loading of damaged coal
Initial damage from engineering disturbances in deep coal mining degrades mechanical properties and heightens dynamic-hazard risks, challenging conventional monitoring. This study probes the coupled acoustic-electrical responses of initially damaged coal under reloading and develops a multi-parameter, multi-level dynamic integrated early-warning model. Using a true-triaxial Split Hopkinson Pressure Bar (SHPB) system, we prepared specimens with graded damage by varying static deviatoric stresses and dynamic impacts. Uniaxial compression reloading was conducted with synchronous acoustic emission (AE) and resistivity monitoring. Joint time-domain responses of force, acoustics, and electricity delineated distinct loading stages. Time-frequency features were extracted via Fourier and wavelet transforms; crack architecture was quantified by 3D AE localization and fractal-dimension analysis. Initial damage markedly reduced load-bearing capacity. Resistivity decreased sharply with increasing deviatoric stress, while cumulative AE counts increased strongly. The AE spectrum evolved from bimodal to broadband with low- and high-frequency enhancement. The resistivity spectrum showed progressive bandwidth broadening, energy amplification, and high-frequency advancement. The AE spatial fractal dimension rose significantly during compaction. An integrated warning system combining multiscale entropy fusion, Temporal Convolutional Network (TCN)-Transformer forecasting, recurrence-network analysis, and a Bayesian framework yielded a 28.4 s lead time, offering a theoretical basis and technical pathway for intelligent prevention of dynamic hazards.
Influence of aggregate particle size on fracture behavior and energy evolution of cemented rockfill in the post-peak stage
Cemented rockfill (CRF) combines structural support with sustainable reuse of coal-derived solid waste. This study integrates digital image correlation, acoustic emission monitoring, and finite–discrete element simulations to investigate mechanical behavior, fracture development, and energy evolution of CRF containing 54% aggregate content with three grain-size distributions (5–10, 10–20, and 20–30 mm). Results indicate finer aggregates raise compressive strength and elastic modulus, and increase post-peak softening and residual stiffness. Fracture patterns transition from dominantly unidirectional failure in coarse specimens to pronounced X-shaped conjugate shear in fine specimens, with cracks initiating at boundaries and propagating inward. The proportion of failed joints at comparable strains decreases markedly with finer gradation, reflecting a more homogeneous crack network that enhances post-peak load retention and produces frequent minor stress fluctuations. Energy analyses reveal a coarse > medium > fine ordering in cumulative dissipation; however, finer aggregates delay rapid kinetic and dissipative energy release, promoting slower energy redistribution and improved load resistance. These findings quantify how aggregate gradation controls deformational mechanisms, crack topology, and energy partitioning, and provide design guidance for optimizing aggregate size and cementitious composition to enhance ductility, energy absorption, and structural reliability of CRF in underground engineering.