Created by Luisa Herrmann and Brenton Crawford
Photographs of the core are taken pretty much whenever, and wherever, core is drilled. Core imagery is information dense and contains a record of valuable rock texture and mineralogical information. Significantly, this is the exact type of information lacking in other downhole datasets such as geochemistry and downhole geophysics. It is inexpensive and not time consuming to acquire, and it represents one of the most ubiquitous geological datasets available.
At Datarock, we have built a Platform that can extract high quality consistent data from traditional core photography using machine learning. Commonly produced datasets include: lithology, alteration, textural analysis, vein analysis, sulphides analysis, RQD, fracture frequency, joint set number and orientation to name a few!
Common issues with historical core photos
One of the key things that can destroy value in the core photography is poor image quality and mistakes made in setting up the core tray for imaging.
We want you to discover the possibility of our powerful workflows and maximise the information you can get out of your imagery! … So, in this blog post we will give you practical tips on how to set up and take the perfect core photo that can maximise the upside for machine learning and automation.
Setting up for the perfect core photograph
A big part of getting the perfect core photo is in the set up and preparation before you take the image. Think of the 3 C’s when setting up your core box Curated, Consistent and Clear!
Tips for setting up your core box correctly
- A physical guide in the core tray will help ensure the image is in the correct consistent position for each image
- Writing the Start and End depths, HoleID and Box Number at the start and end of the core tray allows us to check digital metadata and ensure the image is in the correct orientation and order for analysis
- Curate the rock inside the core tray and make sure it’s clean and in its correct position. Put coarse broken material back together and push finer grain material together to make it represent the correct in-ground length where possible. Don’t push fractured rock together too tightly, leave at least 1 mm of gap so that the fracture position can be detected.
- Core blocks should be clean, neatly written and in the correct orientation (i.e. upright), these can be read by automated systems such as Datarock.
- Automatic geotechnical measurement relies on driller break marks to be clearly annotated with a mark such as an “X”, marks such as these should be in a unique and consistent colour.
- Depth marks on the core should be clearly written as these can be automatically read and used in depth registration, like most other marks they should be a unique and consistent colour.
- A colour calibration chart should be used in the image It can help to track any changes that might occur in the photographic process and also allow for colours to be adjusted/corrected in post processing if required.
- Orientation lines (where present) should be marked clearly on the core in a unique and consistent colour with downhole tick marks on each fragment of rock. The line should be positioned so it is along the bottom of the core so it minimises how much of the rock texture is obscured.
- When taking wet photographs the core should be sprayed with a fine mist that prevents large drops forming. Core should be consistently sprayed with no dry patches.
Aside from the physical core box set up, other important aspects include the collection of metadata in a digital format (i.e HoleID, Box Number and Depths) and the addition of machine readable (barcodes or QR codes) core blocks and core trays.
The Timing of the Core photograph
If the aim is to maximise the ability to extract geotechnical and geological data from the imagery. The best time to take the core photo is after depth marks (metres of feet) and driller break marks are added, but before significant annotation has been applied, as it can obscure the texture of the rock and complicate the machine learning process.
This doesn’t mean that additional photographs of the core couldn’t or shouldn’t be taken. In a perfect world a photograph would be taken at the drill rig in its most undisturbed state, the coreshed before logging, the coreshed after all annotation and again post sampling. This would represent a chain of custody approach to the life cycle of the core tray.
DO’s and DON’Ts of taking a high quality core photograph
DO’s
- Lighting for the photograph should come from a controlled source, inside a light blocking shroud
- Lighting should be diffuse and at a high angle to the core to reduce reflections from wet core
- Set up the camera with the correct settings and do not change them, do not use Automatic – this will result in variable camera settings for each photo ruining consistency
- One core box per photo – if the box is small then move the camera closer to improve the resolution, make sure the entire box is captured.
- A modern DSLR with 30-60 MP this should result in a core photo with resolution between 100-150 microns which would be the current image standard resolution for a modern DSLR camera.
- Ensure consistent camera settings and clean the lens regularly with a lens cleaning kit
- A light source with the highest possible colour rendering index (CRI)
- Dull coloured, non-reflective core boxes
- Camera should be held directly above the box in a fixed frame
DONT’s
- Taking core photos outside using the Sun as a light source
- Illuminating the core from directly above without a light diffuser – this will result in strong reflections
- Using the camera flash particularly without a light diffuser – this will result in reflections on wet core
- Core shed fluorescent lighting – commonly has a low CRI index
- Steel core boxes – these can produce reflections
