Can Hyperspectral Imaging and Neural Network Classification Be Used for Ore Grade Discrimination at the Point of Excavation?

Abstract

This work determines whether hyperspectral imaging is suitable for discriminating ore from waste at the point of excavation. A prototype scanning system was developed for this study. This system combined hyperspectral cameras and a three-dimensional LiDAR, mounted on a pan-tilt head, and a positioning system which determined the spatial location of the resultant hyperspectral data cube. This system was used to obtain scans both in the laboratory and at a gold mine in Western Australia. Samples from this mine site were assayed to determine their gold concentration and were scanned using the hyperspectral apparatus in the laboratory to create a library of labelled reference spectra. This library was used as (i) the reference set for spectral angle mapper classification and (ii) a training set for a convolutional neural network classifier. Both classification approaches were found to classify ore and waste on the scanned face with good accuracy when compared to the mine geological model. Greater resolution on the classification of ore grade quality was compromised by the quality and quantity of training data. The work provides evidence that an excavator-mounted hyperspectral system could be used to guide a human or autonomous excavator operator to selectively dig ore and minimise dilution.

Unsupervised ore/waste classification on open-cut mine faces using close-range hyperspectral data

“…Krupnik (2019) [41] have done a more detailed literature review of studies involving close-range hyperspectral sensing for mining applications. In a project funded by the Minerals Research Institute of WA, Austin et al [42] used hyperspectral imaging to determine ore grade at a mine face in real-time at the point of excavation [43]. This work demonstrated a capability which is considered essential in the development of autonomous mining excavators.…”

Hyperspectral Imaging Algorithms and Applications: A Review

“…In [11], the authors conducted a study to assess the feasibility of using HSI to distinguish ore from waste during excavation in mining operations. They developed a prototype system that combined hyperspectral cameras and LiDAR to capture spatially located hyperspectral data cubes.…”

Onsite Characterisation of Composite Laterite Blocks by Using Novel Sensing Technologies—An Experimental Approach

“…On the other hand, ore discrimination could be achieved by estimating the mineralogical composition using multispectral or hyperspectral images. In this case, the difficulty concerns the need for verification of the relationship between the ore grade and the spectral features of the minerals of the ore [7]. On-site analysis takes place during or after the exploration or the production phase of the deposit.…”

“…In particular, this study's novelty lies in the development of a custom pLIBS scanner that ensures precise instrument movement and the use of commercially available solutions to accurately measure rock properties. difficulty concerns the need for verification of the relationship between the ore grade and the spectral features of the minerals of the ore [7].…”