Key Takeaways

  • Fracture spacing/frequency logging is one of the best-validated AI applications on the whole poster — it has a published head-to-head comparison against expert human logging.
  • Datarock Core (part of IMDEX) uses a neural network to detect and classify fractures directly from core photography.
  • The Carrapateena copper-gold mine study specifically validated this against expert geotechnical logging — a real deposit, real comparison, not a synthetic benchmark.
  • This is a “Yes” — the clearest example on the whole list of AI matching (not just approximating) expert human judgment in a specific geotechnical task.

TL;DR

Run core photography through Datarock Core’s neural-network fracture detection model to get automated fracture spacing/frequency logs that have been validated against expert human geotechnical logging in a real copper-gold mine study.

How Do I Automate Fracture Spacing Logging With AI?

Fracture spacing and frequency (feeding into geotechnical parameters like RQD and FF) is traditionally one of the most labor-intensive and subjective parts of geotechnical core logging — a geologist manually counts and classifies fractures along the core, and different loggers can produce meaningfully different results on the same core.

Datarock Core tackles this with a neural network trained to detect and classify fractures directly from standard RGB core photography, automatically generating fracture frequency and spacing data at a consistency and speed no manual process can match. What sets this apart from a lot of “AI in mining” claims is the evidence: Datarock published a direct comparison of its AI fracture analysis against expert geotechnical logging at the Carrapateena copper-gold mine in South Australia, and the AI output held up against the human benchmark on a real operating deposit, not a synthetic test set.

The practical workflow: core photography (ideally captured with consistent lighting and resolution — see the Core Photography post) feeds into Datarock’s model, which returns fracture locations, spacing, and classification along the core length. Geotechnical engineers then use that output to compute standard indices like RQD and Fracture Frequency for the geotechnical block model, with the option to spot-check or override the AI classification where core condition is ambiguous (e.g. heavily broken or altered zones).

Try It With Geocluster

If you’re evaluating whether AI fracture logging will hold up for your specific rock mass and alteration style — porphyry copper systems can be quite different from Carrapateena’s iron-oxide copper-gold setting — that’s a comparative literature question Geocluster can help you work through quickly.