Close-range, ground-based hyperspectral imaging for mining applications at various scales: Review and case studies

Detailed mapping of mineral phases at centimeter scale can be useful for geological investigation, including resource exploration. This work reviews case histories of ground-based close-range hyperspectral imaging for mining applications. Studies of various economic deposits are discussed, as well as techniques used for data correction, integration with other datasets, and validation of spectral mapping results using geochemical techniques. Machine learning algorithms suggested for automation of the mining workflow are reviewed, as well as systems for environmental monitoring such as gas leak detection. Three new case studies that use a ground-based hyperspectral scanning system with sensors collecting data in the Visible Near-Infrared and Short-Wave Infrared portions of the electromagnetic spectrum in active and abandoned mines are presented. Vertical exposures in a Carlin Style sediment-hosted gold deposit, an active Cu-Au-Mo mine, and an active asphalt quarry are studied to produce images that delineate the extent of alteration minerals at centimeter scale to demonstrate an efficient method of outcrop characterization, which increases understanding of petrogenesis for mining applications. In the Carlin-style gold deposit, clay, iron oxide, carbonate, and jarosite minerals were mapped. In the copper porphyry deposit, different phases of alteration are delineated, some of which correspond to greater occurrence of ore deposits. A limestone quarry was also imaged, which contains bitumen deposits used for road paving aggregate. Review of current literature suggests use of this technology for automation of mining activities, thus reducing physical risk for workers in evaluating vertical mine faces.