Digifrac

Published: 19 May 2025| Version 1 | DOI: 10.17632/9ynpd9956w.1
Contributors:
Bin Wang, Lianhe Sun, Bing Yang

Description

We propose DigiFrac, a Python-based package developed for processing CT scan data to support 3D fracture extraction, complexity analysis, aperture distribution statistics, and orientation calculation. Initially, a random forest algorithm segments the CT data to extract the 3D fracture geometry. Based on this geometry, key parameters—such as fracture volume, fractal dimension, and H-index—are computed. A slice-by-slice thinning algorithm is then applied to extract the fracture skeleton, while the local thickness method is used to determine fracture aperture distribution. Additionally, DigiFrac calculates fracture orientations from the skeleton’s 3D point cloud, enabling detailed, quantitative characterization of complex fracture systems.

Files

Institutions

  • China University of Petroleum Beijing

Categories

Computed Tomography, Rock Fracture

Licence