New vision for drone survey technology
From mining, to ecology and archaeology — smart, drone-mounted hyperspectral camera systems offer a broad landscape of remote sensing possibilities.
Drones with miniaturized, hyperspectral cameras are being used to capture new layers of data that the human eye can’t see, for a broad range of research and industrial applications.
While much of the sun’s warming infrared rays are absorbed by the plants, rocks, water or other objects they strike, some wavelengths get reflected. Different substances reflect different characteristic wavelengths, creating an optical fingerprint of the emitting object. The hyperspectral imaging on these drones captures this infrared light and feeds it back to experts for analysis.
A drone-mounted hyperspectral imaging camera might be used to measure anything from the health of individual plants in a forest or field, to the chemical make-up of a metal-rich ore (see video, below). “What’s really exciting about this technology is its very broad application,” says Steven Micklethwaite, an associate professor from the Sustainable Minerals Institute at The University of Queensland.
Associate Professor Steven Micklethwaite.
Associate Professor Steven Micklethwaite.
Each month, Research News uses the latest findings to help explain the issues facing your community.
Vast data stream
Although hyperspectral imaging cameras have previously been mounted on drones, first-generation systems had significant limitations. Hyperspectral cameras capture a vast data stream which must be calibrated, corrected and processed, a laborious and specialized task.
“You have to account for topography, moisture in the air, shifting sunlight levels from moment to moment – there’s a lot of correction to do before you can even begin to process the data,” Associate Professor Micklethwaite says. “It can take days to weeks to extract information that researchers and industry can use. We're trying to accelerate that, to do it in within four hours or less.”
That advance will partly come from improved hardware. Associate Professor Micklethwaite and his team, as members of the European Union-funded m4mining research consortium, are pursuing a range of approaches from boosting onboard processing power, to incorporating a lidar laser scanner to precisely capture topographic data alongside the hyperspectral data — all while trimming weight to extend drone runtime.
Whereas previous hyperspectral drone cameras only pointed downwards, the new cameras will also be able to tilt, for example, to examine vertical rockfaces for mineral extraction.
Associate Professor Steven Micklethwaite discusses drone sensing, data analytics and stewardship as national Earth-observation infrastructure. Credit: AuScope
Mining and beyond
Associate Professor Micklethwaite and the team are excited for the possibility of capturing hyperspectral data on ore minerals before it is even extracted, or contaminants in mine waste. These insights could enable on the fly process optimization to maximize metal recovery and the management of mine waste for environmental hazards. “That same equipment could also solve critical research questions related to biodiversity, coastal monitoring, agriculture or archaeology,” he says.
The hardware is only part of the story. The team is working hard to improve software to automate and integrate data calibration, correction and processing steps. “Currently you have to do all those things separately, and it’s quite a manual process requiring a lot of expert knowledge,” Associate Professor Micklethwaite says. The team aims to develop automated systems that calibrate and correct multiple parameters simultaneously. “There are a lot of moving parts, and we want to develop software to bring it all together into one package.”
Associate Professor Micklethwaite with a drone in the field.
Associate Professor Micklethwaite with a drone in the field.
The smarter the software, the less demand for specialist expertise. “We’re aiming to deliver the data near real-time and in a way that doesn't require researchers or industry professionals to reskill in order to engage with it,” Associate Professor Micklethwaite says.
From ecologists to archaeologists, researchers could just pick up and use the technology as a powerful tool opening new avenues in their research.
M4Mining is funded by the European Union’s Horizon Europe programme under Grant Agreement ID 101091462. For more information, visit M4Mining. Drone infrastructure and sensing capabilities at UQ are enabled by AuScope and the Australian Government via the National Collaborative Research Infrastructure Strategy (NCRIS).
This article was written by Nature on behalf of The University of Queensland. You can view the original article here.
Built with Shorthand