Leveraging generative AI and hyperspectral drill-core imaging for fully automated mineral mapping of unconventional reservoir
Bibliographic record
Abstract
The acquisition of hyperspectral imagery on core samples significantly enhances the ability of geoscientists to conduct preliminary reservoir characterization particularly for unconventional explorations. Hyperspectral analysis is particularly powerful in providing qualitative and quantitative mineral distribution maps on geological samples. Information about mineral distribution on unconventional shale cores (e.g., clay-dominated vs calcite-dominated) may aid in determining zones of sweet spot for hydraulic fracturing. Conventionally, the mapping of mineralogy in hyperspectral imagery requires considerable human involvement, ranging from the extraction of mineral endmembers, process of defining the representative reflectance spectra signature of every minerals in the hyperspectral dataset, to the selection of parameters for classification algorithms. We used a generative deep learning approach with modified Deep Embedded Clustering (DEC) and variational autoencoders (VAE) to automate mineral classification in hyperspectral imagery of Devonian unconventional reservoir cores from the Western Canada Sedimentary Basin. The cores are characterized by nodular carbonates and Si- and clay-rich shales with interbedded detrital carbonates, covering the upper portion of the Waterways Formation, the entire Majeau Lake and Duvernay formations, and a portion of the lower Ireton Formation. Our study demonstrates that the proposed generative AI approach is effective in qualitatively and quantitatively identifying key mineral endmembers, including calcite, non-clay silica-rich minerals, and clay minerals (illite and muscovite) and their mixtures. In addition, the algorithm is also able to extract and identify the spectral variation of the silica-rich minerals based on its organic content. The VAE+DEC algorithm was able to capture not only the thin interbedded layers but also its complex mineralogy that aligns well with findings from previous research on the studied formations. In addition, the spatial distribution of organic content variation is also identified correctly and matched with the laboratory measurements. Therefore, the proposed workflow emerges as a promising, less labor-intensive alternative for lithological mapping and brief rock properties analysis from hyperspectral datasets, offering potential further applications in unconventional reservoir drill core data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".