Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 25, 2026Applied SciencesOpen Access

Attention-Guided Semantic Segmentation and Scan-to-Model Geometric Reconstruction of Underground Tunnels from Mobile Laser Scanning

View Full Paper
Ask AI
Bookmark
Share

Authors

YHYingjia HuangJYJiang YeXLXiaohui Li

Discussion

Loading...

Member takes

Overview

Demonstrates an automated framework for tunnel segmentation and geometric modeling, suggesting improvements in mining infrastructure evaluation.

Key Points

  • The aim is to automate the interpretation of point clouds into engineering models for underground environments.
  • Developed a hierarchical deep learning network for semantic segmentation of point clouds.
  • Implemented a context-aware sampling strategy with an expanded receptive field for improved continuity.
  • Utilized a composite Focal-Dice loss function to address class imbalance in training.
  • Conducted experiments on a dataset of 16 mine tunnels for validation.
  • Established a geometric modeling pipeline using curvature-guided filtering.
  • Achieved a mean Intersection over Union (mIoU) of 85.15% (±0.29%).
  • Obtained an Overall Accuracy (OA) of 95.13% (±0.13%).
  • Filtered Root Mean Square Error (RMSE) of 4.96 ± 0.48 cm for reconstructed profiles.

Cite This Study

Huang et al. (2026) studied this question.

synapsesocial.com/papers/69c37bd4b34aaaeb1a67ea81https://doi.org/10.3390/app16063042
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Tunnel scanner: Geometry-informed synthetic point cloud generation and transfer learning for tunnel segmentation2026 · 1 citations
  2. 2Group-aware and position-interactive learning for point cloud robust segmentation in subterranean environments2026
  3. 3Lidar-based Simultaneous Localization and Mapping with Connecting Traverse Constraints for Underground Mine Mapping2026
  4. 4Parallel Dual-Level Attention-Enhanced Semantic Segmentation of 3D Point Clouds in Tunnel Construction2026
  5. 5Semantic segmentation of large-scale segmental lining point clouds using 3D deep learning2024 · 2 citations