Probabilistic clutter maps of forested terrain from airborne LiDAR point clouds

Detection from airborne sensors of near-ground objects occluded by above-ground vegetation is not usually straightforward. Our hypothesis is that the probability of obstruction due to objects above ground at any location in the forest environment can be estimated with measurable uncertainty from airborne lidar data. The essence of our approach is to develop a data-driven learning scheme that creates 2D probability maps for obstructions at the study site. The result shows the effectiveness of the newly developed individual tree detection algorithm (with the accuracy index of 77.1%, tested using ground surveys) and also the usefulness of the clutter and uncertainty maps in the prediction of line-of-sight visibility, mobility and above-ground forest biomass.