Today's mine planning and open pit mining operations rely on manual interpretation of geology, visual and geophysical logging of drill cores, low-resolution surveys, multispectral satellite data, and sampling of rock properties. These methods are time-consuming and don't offer a complete picture of mineral distribution in active mines and tailing surfaces.
Spectral remote sensing which uses the unique signatures of materials in reflected light enables mineral mapping at various scales, including site-specific. However, this technique is currently used by only a few mining service providers. None include high-resolution mine-face scanning, correction, and data calibration from drone platforms, and no commercial solutions offer real-time processing results.
The results achieved should be compared to this background. Within the first 18 months of the project, m4mining has already achieved several results that are beyond the current state of the art. Notably, we have developed a functional drone system equipped with sensors for imaging/mineral classification and 3D surface mapping. While the individual components are commercially available, the integration for side-viewing data acquisition on steep surfaces and real-time analysis required research and custom development. Furthermore, we have developed a three-dimensional surface mapping system to accurately map the mineral distribution of commonly steep and complex faces of open pit benches, stock and mine waste piles, as well as tailings dams.
One reason that the reflectance imaging technique for material characterization hasn't been widely adopted in mining operations is the prohibitively long and complex data processing. To address this, we have developed software that delivers near real-time results (same day) without requiring dedicated experts for data interpretation, representing another advancement beyond the current state of the art.
Finally, we have identified suitable algorithms for real-time and near-real-time mineral classification, capable of handling the scale differences between drone and satellite imagery. This achievement contributes to establishing a standard practice for analyzing such data in the mining industry.