A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
A human-machine interaction method for rock discontinuities mapping by three-dimensional point clouds with noises
Rock discontinuities control rock mechanical behaviors and significantly influence the stability of rock masses. However, existing discontinuity mapping algorithms are susceptible to noise, and the calculation results cannot be fed back to users timely. To address this issue, we proposed a human-machine interaction (HMI) method for discontinuity mapping. Users can help the algorithm identify the noise and make real-time result judgments and parameter adjustments. For this, a regular cube was selected to illustrate the workflows: (1) point cloud was acquired using remote sensing; (2) the HMI method was employed to select reference points and angle thresholds to detect group discontinuity; (3) individual discontinuities were extracted from the group discontinuity using a density-based cluster algorithm; and (4) the orientation of each discontinuity was measured based on a plane fitting algorithm. The method was applied to a well-studied highway road cut and a complex natural slope. The consistency of the computational results with field measurements demonstrates its good accuracy, and the average error in the dip direction and dip angle for both cases was less than 3°. Finally, the computational time of the proposed method was compared with two other popular algorithms, and the reduction in computational time by tens of times proves its high computational efficiency. This method provides geologists and geological engineers with a new idea to map rapidly and accurately rock structures under large amounts of noises or unclear features.
A human-machine interaction method for rock discontinuities mapping by three-dimensional point clouds with noises
Rock discontinuities control rock mechanical behaviors and significantly influence the stability of rock masses. However, existing discontinuity mapping algorithms are susceptible to noise, and the calculation results cannot be fed back to users timely. To address this issue, we proposed a human-machine interaction (HMI) method for discontinuity mapping. Users can help the algorithm identify the noise and make real-time result judgments and parameter adjustments. For this, a regular cube was selected to illustrate the workflows: (1) point cloud was acquired using remote sensing; (2) the HMI method was employed to select reference points and angle thresholds to detect group discontinuity; (3) individual discontinuities were extracted from the group discontinuity using a density-based cluster algorithm; and (4) the orientation of each discontinuity was measured based on a plane fitting algorithm. The method was applied to a well-studied highway road cut and a complex natural slope. The consistency of the computational results with field measurements demonstrates its good accuracy, and the average error in the dip direction and dip angle for both cases was less than 3°. Finally, the computational time of the proposed method was compared with two other popular algorithms, and the reduction in computational time by tens of times proves its high computational efficiency. This method provides geologists and geological engineers with a new idea to map rapidly and accurately rock structures under large amounts of noises or unclear features.
A human-machine interaction method for rock discontinuities mapping by three-dimensional point clouds with noises
Qian Chen (author) / Yunfeng Ge (author) / Changdong Li (author) / Huiming Tang (author) / Geng Liu (author) / Weixiang Chen (author)
2025
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
| Elsevier | 2025
Rock Discontinuities Identification from 3D Point Clouds Using Artificial Neural Network
| Online Contents | 2022
An efficient method for extracting and clustering rock mass discontinuities from 3D point clouds
| Springer Verlag | 2023
| Elsevier | 2024