Machine Learning for Mineral Identification and Ore Estimation from Hyperspectral Imagery in Tin–Tungsten Deposits: Simulation under Indoor Conditions

Abstract

This study aims to assess the feasibility of delineating and identifying mineral ores from hyperspectral images of tin–tungsten mine excavation faces using machine learning classification. We compiled a set of hand samples of minerals of interest from a tin–tungsten mine and analyzed two types of hyperspectral images: (1) images acquired with a laboratory set-up under close-to-optimal conditions, and (2) a scan of a simulated mine face using a field set-up, under conditions closer to those in the gallery. We have analyzed the following minerals: cassiterite (tin ore), wolframite (tungsten ore), chalcopyrite, malachite, muscovite, and quartz. Classification (Linear Discriminant Analysis, Singular Vector Machines and Random Forest) of laboratory spectra had a very high overall accuracy rate (98%), slightly lower if the 450–950 nm and 950–1650 nm ranges are considered independently, and much lower (74.5%) for simulated conventional RGB imagery. Classification accuracy for the simulation was lower than in the laboratory but still high (85%), likely a consequence of the lower spatial resolution. All three classification methods performed similarly in this case, with Random Forest producing results of slightly higher accuracy. The user’s accuracy for wolframite was 85%, but cassiterite was often confused with wolframite (user’s accuracy: 70%). A lumped ore category achieved 94.9% user’s accuracy. Our study confirms the suitability of hyperspectral imaging to record the spatial distribution of ore mineralization in progressing tungsten–tin mine faces.

Case Study on the Fitting Method of Typical Objects

“…The evaluation indicators are shown in Table 8. Tables [5][6][7] show that the target polynomial fitting of mica schist, grass, loam, asphalt, and Jasper Ridge gravel has little difference with Fourier fitting, and the indicators are relatively close.…”

“…) are energy constraints. Y j is the estimated value calculated by Equation (5) When Equation (1) is not discretized, calculate the average energy mean of the target spectrum in the range of 400-900 nm. T and T are used as constraints to solve the overall mean value of the discrete spectrum.…”

“…The target spectrum curve fitting evaluation index is shown in Tables 5-7. Tables [5][6][7] show that when there is an error, the discrete value calculated by AX=B is fitted. Mica schist, grass, Jasper Ridge gravel, and asphalt have the best fitting results by polynomial fitting, while copper metal and loam have the best fitting precision by Fourier fitting.…”

“…In recent years, hyperspectral technology has been widely used in agriculture [1][2][3][4], resource exploration [5][6][7][8], oceanic studies [9], and environmental research [10]. Spectrometers are gradually becoming intelligent, miniaturized, and lightweight.…”

Hierarchical Unsupervised Partitioning of Large Size Data and Its Application to Hyperspectral Images

“…Due to this richness of information, which allows for better discrimination of objects, interest in hyperspectral images has increased during the last years in many application fields. These fields include geology [1], medicine [2][3][4][5][6][7], industrial production [8,9], safety [10], and the environment [11][12][13][14][15][16][17][18][19][20]. In this latter field, hyperspectral imagery has been of great interest and several applications have been treated.…”