To document

Abstract

Detailed forest inventories are critical for sustainable and flexible management of forest resources, to conserve various ecosystem services. Modern airborne laser scanners deliver high-density point clouds with great potential for fine-scale forest inventory and analysis, but automatically partitioning those point clouds into meaningful entities like individual trees or tree components remains a challenge. The present study aims to fill this gap and introduces a deep learning framework, termed ForAINet, that is able to perform such a segmentation across diverse forest types and geographic regions. From the segmented data, we then derive relevant biophysical parameters of individual trees as well as stands. The system has been tested on FOR-Instance, a dataset of point clouds that have been acquired in five different countries using surveying drones. The segmentation back-end achieves over 85% F-score for individual trees, respectively over 73% mean IoU across five semantic categories: ground, low vegetation, stems, live branches and dead branches. Building on the segmentation results our pipeline then densely calculates biophysical features of each individual tree (height, crown diameter, crown volume, DBH, and location) and properties per stand (digital terrain model and stand density). Especially crown-related features are in most cases retrieved with high accuracy, whereas the estimates for DBH and location are less reliable, due to the airborne scanning setup.

Abstract

Mapping individual tree quality parameters from high-density LiDAR point clouds is an important step towards improved forest inventories. We present a novel machine learning-based workflow that uses individual tree point clouds from drone laser scanning to predict wood quality indicators in standing trees. Unlike object reconstruction methods, our approach is based on simple metrics computed on vertical slices that summarize information on point distances, angles, and geometric attributes of the space between and around the points. Our models use these slice metrics as predictors and achieve high accuracy for predicting the diameter of the largest branch per log (DLBs) and stem diameter at different heights (DS) from survey-grade drone laser scans. We show that our models are also robust and accurate when tested on suboptimal versions of the data generated by reductions in the number of points or emulations of suboptimal single-tree segmentation scenarios. Our approach provides a simple, clear, and scalable solution that can be adapted to different situations both for research and more operational mapping.