EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Point cloud feature stitching and fine-grained modeling of complex roadway based on LiDAR

Mingchao Du, Changjiang Yi, Kun Zhang, Yuxia Li, Liangsong Huang, Tongzhu Yu, Daixiang Zhang, Jian Ma, Shibin Yang

AIP Advances · 01.11.2026 · 16 (11)

DOI: 10.1063/5.0332873 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AIP Advances, online erschienen am 20.11.2026. Verfasst von Mingchao Du und 8 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Major Innovation Projects in Shandong Province, Natural Science Foundation of Shandong Province, Technology Innovation Guidance of Shandong Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the advancement of intelligent coal mining, higher demands have been placed on the modeling accuracy of 3D maps of coal mine tunnels. To address issues such as low point cloud registration accuracy and loss of modeling details in complex tunnel environments, this paper proposes a method for feature-based point cloud registration and high-resolution modeling in complex tunnels. Two LiDAR systems with different thread counts were selected as 3D data acquisition devices. First, radius filtering and voxel filtering are applied to the point cloud to thin it out, remove large-scale noise, and compress the point cloud data. Second, the point cloud is processed using 3D-SIFT feature point detection and the FPFH descriptor to extract tunnel contour features. Subsequently, point cloud registration is performed using the improved RANSAC algorithm for coarse registration and the improved LM-ICP algorithm for fine registration. Finally, 3D contour reconstruction experiments were conducted on rectangular and arched simulated tunnels. The experimental results show that the improved LM-ICP fine registration algorithm achieved a point cloud registration error of ≤2.1 mm in the simulated tunnels, enabling detailed tunnel modeling. This method improves the RANSAC algorithm through geometric constraints on polygon side lengths, addressing the issue of misalignment during coarse registration in noisy tunnel environments, and provides technical support for environmental perception of mobile equipment in underground coal mines.

Förderung

  • Major Innovation Projects in Shandong Province
  • Natural Science Foundation of Shandong Province
  • Technology Innovation Guidance of Shandong Province
  • Development Plan for Youth Innovation Teams in Higher Education Institutions in Shandong Province

Zitieren

APA

Du, M., Yi, C., Zhang, K., Li, Y., Huang, L., Yu, T., Zhang, D., Ma, J., & Yang, S. (2026). Point cloud feature stitching and fine-grained modeling of complex roadway based on LiDAR. AIP Advances, 16(11). https://doi.org/10.1063/5.0332873

Vancouver

Du M, Yi C, Zhang K, Li Y, Huang L, Yu T, et al. Point cloud feature stitching and fine-grained modeling of complex roadway based on LiDAR. AIP Advances. 2026;16(11). doi:10.1063/5.0332873

BibTeX
@article{Du2026,
  author = {Du, Mingchao and Yi, Changjiang and Zhang, Kun and Li, Yuxia and Huang, Liangsong and Yu, Tongzhu and Zhang, Daixiang and Ma, Jian and Yang, Shibin},
  title = {Point cloud feature stitching and fine-grained modeling of complex roadway based on LiDAR},
  journal = {AIP Advances},
  year = {2026},
  volume = {16},
  number = {11},
  doi = {10.1063/5.0332873}
}
Kostenlos starten →