Modelling Elastic Properties of Clastic Rocks from Microtomographic Images Using Multi-Mineral Segmentation and Machine Learning
Modelling elastic properties from micro-CT images of rocks is essential for geophysical characterisation of the subsurface. This is achieved through an advanced physics-based multi-mineral image segmentation workflow, which is then automated using machine learning. The effects of intergranular conta...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
Curtin University
2022
|
| Online Access: | http://hdl.handle.net/20.500.11937/88666 |
| Summary: | Modelling elastic properties from micro-CT images of rocks is essential for geophysical characterisation of the subsurface. This is achieved through an advanced physics-based multi-mineral image segmentation workflow, which is then automated using machine learning. The effects of intergranular contacts that are below the micro-CT resolution are modelled by a workflow that extracts their elastic properties from rock microstructure and ultrasonic measurements. I also developed a workflow that successfully detects pressure-induced deformation in micro-CT images. |
|---|