Extrapolating rock physicochemical properties from discrete sample data using computer vision
Lewis J C Grant, Jennifer Lington, Elliot J Carter, Damon A H Teagle, Thomas M Belgrano, Rosalind M Coggon · 2026
Read Extrapolating rock physicochemical properties from discrete sample data using computer vision by Lewis J C Grant, Jennifer Lington, Elliot J Carter, Damon A H Teagle, Thomas M Belgrano, Rosalind M Coggon (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13282442.txt.
Description
Lewis J C Grant and colleagues published a 2026 study using computer vision to stretch a handful of laboratory measurements across a hundred meters of drilled ocean crust. Geologists normally describe core visually and sample sparsely, because no expert can annotate every centimeter. Working on a carbonate-cemented breccia core from beneath the seafloor, the team trained deep learning classifiers on the core imagery to label rock types. They then projected measured bulk density and carbon content onto the whole core at quarter-square-centimeter resolution.
The density record improved on the instrument measurement it was checked against. The carbon estimate underlined how much carbon this broken, cemented seafloor rock locks away, which matters for global carbon accounting. The authors pitch the framework as scalable to the world's existing core repositories: archives of already-drilled material could be re-audited automatically, turning stored imagery into continuous quantitative records.
Catalog record
| CREATOR | Lewis J C Grant, Jennifer Lington, Elliot J Carter, Damon A H Teagle, Thomas M Belgrano, Rosalind M Coggon |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |