Abstract
Digital images obtained from X-ray computed tomography scans are analysed for the estimation of inclusion shape and orientation. Three-dimensional computer imagery and segmentation algorithms are used to visualise and isolate the regions of interest. These regions are then approximated by best-fit ellipsoids and the mean best-fit ellipsoid is used as a measure of preferred inclusion orientation. A Windows program is developed to implement these procedures and results found from both manufactured and natural data are presented. These results show that the radiodensity contrast plays a major role in the ability of the software to isolate inclusions from their matrix and hence determine rock fabric.
| Original language | English |
|---|---|
| Pages (from-to) | 1358-1368 |
| Number of pages | 11 |
| Journal | Computers and Geosciences |
| Volume | 35 |
| Issue number | 6 |
| DOIs | |
| Publication status | Published - Jun 2009 |
Keywords
- CT scanning
- Ellipsoid fitting
- Fabric analysis
- Image segmentation
- Non-invasive object inspection