Articles | Volume 67
https://doi.org/10.5194/adgeo-67-79-2025
https://doi.org/10.5194/adgeo-67-79-2025
03 Dec 2025
 | 03 Dec 2025

Quantifying carbonate microstructure using classical segmentation pipelines for CCS and radioactive waste applications

Wurood S. Alwan, Omar Choudhry, Paul Glover, Louey Tliba, and Richard Collier

Viewed

Total article views: 49 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
32 13 4 49 1 2
  • HTML: 32
  • PDF: 13
  • XML: 4
  • Total: 49
  • BibTeX: 1
  • EndNote: 2
Views and downloads (calculated since 03 Dec 2025)
Cumulative views and downloads (calculated since 03 Dec 2025)

Viewed (geographical distribution)

Total article views: 48 (including HTML, PDF, and XML) Thereof 48 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 06 Dec 2025
Download
Short summary
We mapped pore networks in carbonate scanning-electron-microscope (SEM) images to aid safe CO2 storage and radioactive-waste isolation. An open 29 000 × 23 000-pixel mosaic was split into 100 fully hand-labelled tiles. Eight classic computer-vision pipelines were benchmarked: Watershed gave the best recall–precision balance, and a simple Hybrid-Voting of three filters halved false alarms. The data, masks and code offer an immediate reference set for training and testing future 3-D deep-learning models.
Share