Novel mapping of Scotland’s rainforests using state-of-the-art drone imaging and machine learning

A joint project between the University of Edinburgh’s School of Geosciences and Forestry and Land Scotland (funded by Forestry and Land Scotland), is currently underway exploring the use of ultra-high resolution aerial LiDAR, multispectral and RGB data to map invasive species in Scotland’s rainforests.

landscape

Using our in house capability at Airborne Research and Innovation (www.ed.ac.uk/airborne), UAV flights were carried out in April of this year on the west coast of Scotland at a “leaf off” time such that understory species (rhododendron, holly and regenerating spruce) could be visualised and mapped.   A key objective in this work is to use machine learning classification methods to better differentiate rhododendron from other species, to allow management teams to prioritise action areas for control measures. The analysis phase will continue into the autumn this year, with published results forthcoming.

For more information on the project, please contact Dr Caroline Nichol at caroline.nichol@ed.ac.uk.

This article was published on