Automating Fracture Detection and RQD Calculation in Core Images Using Deep Learning
Bibliographic record
Abstract
In the geological and geotechnical engineering fields, drill cores are a vital source of information. However, manual core logging faces several challenges: it is time-consuming, labor-intensive, and subject to the biases of the geologists conducting the assessments. To address these issues, this study explored the potential of digital core logging automation techniques. We employed a deep learning approach, specifically using the YOLOv8 (You Only Look Once) model, to identify different classes, including clean fractures, intact rocks, depth blocks, broken joints, as well as empty trays within a core. This model achieved an average precision of 89.1% and a recall of 81.3% across all classes, with a mean Average Precision at 50% overlap (mAP50) of 88.2%. This indicated a relatively high level of accuracy in the YOLO model’s performance. Furthermore, the clean fracture class had a precision of 90.6%, a recall of 75.5%, and an mAP50 of 88.4%. The intact rock class showed even higher performance with an mAP50 of 95%, along with a precision of 86.1% and a recall of 91.5%.Additionally, we introduced a post-processing code designed to automate the calculation of Rock Quality Designation (RQD) and fracture count per depth. A comparative analysis revealed that the RQD values estimated by our model displayed an approximate 5% absolute error compared to the ground truth RQD calculations performed on the drill cores This study highlighted the effectiveness of deep learning in improving the accuracy and efficiency of core log analysis, suggesting its potential to revolutionize manual core logging techniques and shape the future of core analysis.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".