This document details Datarock’s Unique Classification products.
Dependent Pipelines
The outputs of the following pipelines are used to determine any Unique Classification:
Pipeline Name |
Pipeline Output Type |
Image Preparation |
Object Detection |
Depth Registration |
Semantic or Instance Segmentation |
Unique Classification |
Classification |
Data Processing
The outputs of the Datarock Unique Classification model when applied over depth registered rows, allow classification labels including dominant label and class probability statistics of these predictions to be calculated.
Depth-registered rows are cut into 5cm squares ('tiles') using customisable clipping and overlap parameters of the row imagery using the following settings.
-
Clip the top and bottom of each row image by 15%
-
This removes influences from the edge of the tray
-
-
Overlap the squares by 20%
-
This allows for capturing geological structure and texture at tile boundaries that could otherwise be unaccounted for.
-
An example of how the squares appear on depth-registered rows is shown below.
Below is an image illustrating the difference between a classification model and its potential outputs against an object detection, semantic segmentation (single-class) and instance segmentation (multi-class) model.
Detection of Unique Classification classes
Several classification classes can be detected by a Unique Classification model based on either manual labelling of 5cm resolution tile imagery during onboarding, sampling for tiles based on geological logs or via UMAP using the textural and colour differences of these tiles. The latter two can assist in further augmenting the training process.
These annotated tiles are split into training and evaluation datasets and the following image shows a compilation of training tiles for each class of a 10-class lithology classification model.
Product Configuration Options
There are no configuration aspects to this product.
Output Intervals
Default interval lengths: raw data is produced at 0.05m scale as well as composited intervals of 0.5m, 1m, 2m, 3m and 5m.
Customisable interval is available: Yes, via uploading an assay or geology logging table to Datarock Customer Success Team.
User Data
User data can not be uploaded to the Platform via CSV at the current time.
The following data is required for customisable intervals (assay or geology logs) to be sent to Datarock Customer Success Team:
-
Hole_ID
-
Depth_From
-
Depth_To
-
Class name (ie. lithology, vein occurrence, alteration)
Data Output
Results from this class of model can be obtained using the Download Square Classification artefacts option from the Actions button in the Model Review tab of Datarock. The available CSV files include the following for the drill hole(s) selected:
-
ProjectID_HoleID_square_classification_composite_Xm.csv
-
ProjectID_HoleID_square_classification_composite_segment_Xm.csv
-
ProjectID_HoleID_square_classification_interval.csv
-
ProjectID_HoleID_square_classification_square.csv
These four CSV files contain the following headers:
Composite Data Files |
ProjectID_HoleID_square_classification_composite_Xm.csv;ProjectID_HoleID_square_classification_composite_custom_intervals_Xm.csv; |
Column Header |
Description |
hole_id |
Customer’s Hole ID |
interval_from |
Start of interval |
interval_to |
End of interval |
dominant_class |
Model predicted outcome of most likely/majority class within selected interval |
length_interval |
Length of tile (5cm) |
length_response |
Length of core detected in 5cm tile |
length_valid |
Length of core detected in 5cm tile used for prediction |
class name |
Certain quantity of columns based on name and number of trained classes. Each row shows proportional prediction percentage of each class for that interval (adds up to 100%). |
|
|
Raw Data Files |
ProjectID_HoleID_square_classification_interval.csvProjectID_HoleID_square_classification_square.csv |
Column Header |
Description |
hole_id |
Customer’s Hole ID |
box_number |
Platform assigned box number reference |
depth_from |
Start of tile interval (metres) |
depth_to |
End of tile interval (metres) |
class |
Class predicted for square interval |
probability |
Statistical confidence of the model in selecting correct class for the square |
inference_timestamp |
Timestamp of when the model calculated the class |
classification_source |
What was used to derive the class prediction (model) |
edited_by |
Name of author of any edits conducted on the class outcome (if applicable) |
version |
Model version identifier |
Product Limitations
Limitations |
Comments |
Reliance of row detection and depth registration |
The Unique Classification model is based on predicting geological features within a 5cm tile cut from row imagery. The dependency on the depth registered rows being identified means if a row is missed by the row model during Image Preparation, the subsequent tiles will not have the classification modelled applied. |
Training is dependant on what can be seen within a tile image |
Datarock’s Unique Classification model relies on classes being predicted using visually identifiable RGB features some of which are too subtle to predict from a photo, in particular if resolution is poor. Tile imagery during the training process also does not provide geological context for individual squares when sampled at random. For example, two red-coloured tiles could both be predicted as “oxide” but one is from 1m depth in a surface hole (weathered rock and therefore true positive) and the other at 1,000m depth (hematite-oxidation in an IOCG and therefore a false positive). Site-based logging can be used to assist in class training of these false positives, however logging is not always at the same resolution as 5cm tiles. Tiles which fall at the end of the rows may only contain a fraction of rock and thus can cause confusion for the model. To minimise any model confusion, these examples are generally trained as an “Other” class along with tiles that contain solely core block or empty tray. |
Training data must be representative of whole area classification is to be applied |
If new imagery or classification class is introduced to the model, the performance may decline as these examples were not trained during onboarding. An initial model evaluation will need to be undertaken to see the suitability of the model in particular against any new imagery. Ideally, a new model version is trained to incorporate the new untrained tiles or class. |
Document Version
Version |
Date |
Author |
Rationale |
1 |
11 August 2023 |
C Brown |
Initial release |