Abstract
Drone based hyperspectral and LiDAR systems for monitoring mining environment
Monitoring of sensitive ecosystems, such as swamp vegetation, in mining areas is an important yet challenging task due to their inherent complexities. Cutting edge sensors mounded on Unmanned Aerial Vehicles (UAVs) or drones are shaping the way we monitor our ecosystems today. It is now possible to use light weight sensors such as hyperspectral and LiDAR on drones to obtain detailed bio-physio-chemical characteristics of the environment at a very fine scale spatial resolution. The mobile system enables to accurately map species distribution, diversity, natural chlorophyll levels, photosynthetic activity, foliage cover, leaf area index and much more. This presentation demonstrates the use of drone based hyperspectral and LiDAR systems to produce detailed map of swamp vegetation in an ecologically complex swamp environment at a mines site. A Fabry–Pérot interferometer (FPI) type lightweight hyperspectral sensor was used to generate geo-spatialised classified raster products unique to the mine environment. In addition, a light-weight LiDAR unit was used to derive structural matrices for species level classification. The preliminary result is encouraging and sets the basis for further research in development of unique data processing algorithms, such as data fusion and machine learning, to improve the mapping accuracies.
Drone based hyperspectral and LiDAR systems for monitoring mining environment
Monitoring of sensitive ecosystems, such as swamp vegetation, in mining areas is an important yet challenging task due to their inherent complexities. Cutting edge sensors mounded on Unmanned Aerial Vehicles (UAVs) or drones are shaping the way we monitor our ecosystems today. It is now possible to use light weight sensors such as hyperspectral and LiDAR on drones to obtain detailed bio-physio-chemical characteristics of the environment at a very fine scale spatial resolution. The mobile system enables to accurately map species distribution, diversity, natural chlorophyll levels, photosynthetic activity, foliage cover, leaf area index and much more. This presentation demonstrates the use of drone based hyperspectral and LiDAR systems to produce detailed map of swamp vegetation in an ecologically complex swamp environment at a mines site. A Fabry–Pérot interferometer (FPI) type lightweight hyperspectral sensor was used to generate geo-spatialised classified raster products unique to the mine environment. In addition, a light-weight LiDAR unit was used to derive structural matrices for species level classification. The preliminary result is encouraging and sets the basis for further research in development of unique data processing algorithms, such as data fusion and machine learning, to improve the mapping accuracies.