UAV-based LiDAR acquisition for the derivation of high-resolution forest and ground information

Abstract

Laser scanning of forested areas helps in analyzing and understanding various aspects of forest conditions, including distribution of plants and trees, height distribution of trees, tree density, size and volume of wood, as well as ground surface properties. However, laser scanning of forest areas is also very challenging for many reasons. The best time for scanning is before trees leaf out in the spring or after trees cast their leaves in autumn before snowfall so an unmanned aerial vehicle (UAV) laser scanner can penetrate the forest from the tops of the trees down to the ground surface. To receive highly accurate laser data and high point density, the flight planning must be adjusted judiciously. Flight planning will be even more complex in steep terrain where the UAV cannot operate at a constant altitude. This paper discusses a UAV-based 3D laser data recording — light detection and ranging (LiDAR) scanning — of a forestry area with high accuracy and point cloud resolution. In addition, the point cloud of airborne laser scanning is compared with local terrestrial laser-scanning results. The forest area consists of mixed forests containing varying tree sizes and branch deformation. This paper summarizes our latest results in UAV-based LiDAR acquisition over a forest area to extract detailed forest and ground information and finds that UAV-based laser scanning is well suited for provision of both high-quality forest structural and terrain elevation information.