· McEwen Mining (TSX:MUX) – Grey Fox deposit area, a part of the Black Fox Complex in the historic Timmins Gold Camp, Canada. 136,000 meters relogged using LithoLens – 3 generations of veins and alterations identified.
· Avalon Investment Holdings (Omai Gold Mines Corp.) (privately held company) – Omai Gold Mine in Guyana, 27,300 meters relogged using LithoLens to identify quartz veining which was believed to be the host of gold mineralization in this paleoprterozoic orogenic gold system in the Guyana shield, which is analogous to the archean deposits of the Canadian shield.
· Cassiar Gold Corp (TSXV:GLDC) – Cassiar Gold Project in northern British Columbia, 10,200 meters relogged using LithoLens for quartz vein detection and alteration associated with mineralization.
· TriStar Gold (TSXV:TSG) – Castelo de Sonhos Gold Project in the southwestern Pará state, Brazil, televiewer – 14,000 meters relogged using LithoLens for cobble detection, which is critical for determining stratigraphic context and location within paleochannel environments and guiding further exploration.
Every mining exploration company is legally required to photograph drillcore boxes as they are collected. These boxes are later visually inspected by geologists, and their core composition intervals manually recorded.
Most deposits have accumulated thousands of core photos, with more images collected with every new drill campaign. While some of the core may be lost, theses photos are stored on various servers but often forgotten. Millions of dollars have been spent collecting these images, only to have them remain underutilized.
We extract untapped value from these core photos by using LithoLens, which is a series of deep learning algorithms to clean and optimize the quality of the existing images, stitch them together into a complete core image, and then automatically recognize and document key geological features within the core. LithoLens uses existing images that would otherwise sit, unexamined, on a server or hard drive.
LithoLens, GoldSpot’s image feature extraction technology, will automatically identify and document key information from historic core photography.
The core relogging workflow utilizes secure, cloud-based data processing and includes two distinct Deep Learning steps.
The first step is automated photo homogenization, where image data is cleaned, processed, extracted and georeferenced.
The second step is deployment of a Deep Learning algorithm that can recognize varying geological intervals – or specific features such as veins – within the core.
This process utilizes existing core images that typically have no relevant purpose, and extracts relevant geological information for consideration and integration.
The machine learning model can be supervised and trained to re-calibrate and then automatically applied to subsequent images.
Once the machine learning model has been trained, it can process 1 m of core in less than 1.5 seconds, converting photos into accurate useable data.
137 km of core was automatically converted to CSV data files in less than 48 hours, once the period of data training had been completed.
This CSV file contains quantified numeric data representing vein density, alteration intensity or rock type, all including holeID and from – to information.
Contact us today to utilize the idle data you already have.