Drone-Based VNIR–SWIR Hyperspectral Imaging for Environmental Monitoring of a Uranium Legacy Mine Site

Citations Over TimeTop 10% of 2025 papers

Abstract

Growing awareness of the environmental cost of mining operations has led to increased research on monitoring and restoring legacy mine sites. Hyperspectral imaging (HSI) has emerged as a valuable tool in the mining life cycle, including post-mining environment. By detecting variations in crystal structure and physicochemical attributes on the surface of materials, HSI provides insights into site environmental and ecological conditions. Here, we explore the capabilities of drone-based HSI for mapping surface patterns related to contamination dispersal in a legacy uranium-rare earth element mine site. Hyperspectral data across the visible to near-infrared (VNIR) and short-wave infrared (SWIR) wavelength ranges (400–2500 nm) were collected over selected areas of the former Mary Kathleen mine site in Queensland, Australia. Analyses were performed using data-driven (Spectral Angle Mapper—SAM) and knowledge-based (Band Ratios—BRs) spectral processing techniques. SAM identifies contamination patterns and differentiates mineral compositions within visually similar areas. However, its accuracy is limited when mapping specific minerals, as most endmembers represent mineral groups or mixtures. BR highlights reactive surfaces and clay mixtures, reinforcing key patterns identified by SAM. The results indicate that drone-based HSI can capture and distinguish complex surface trends, demonstrating the technology’s potential to enhance the assessment and monitoring of environmental conditions at a mine site.

Related Papers

• → Visible to SWIR hyperspectral imaging for produce safety and quality evaluation(2011)31 cited
• → Development of a hyperspectral scene generator(2003)6 cited
• → Detection and correction of spectral and spatial misregistrations for hyperspectral data(2010)4 cited
• → Dimensionality reduction for spatial-spectral target detection on hyperspectral imagery(2018)
• → Lithotype Classification in Geothemal Area by the Use of Hyperspectral Data(2023)