Radiometric features improve leaf and wood segmentation in aerial LiDAR scans
The Role of Radiometric Features in Cross-Site Leaf-Wood Segmentation of LiDAR Point Clouds
Computer Vision and Pattern Recognition
Summary
Identifying tree parts like leaves and wood using 3D laser scans helps estimate how much biomass a tree has without cutting it down. Most methods look only at shapes and ignore extra light intensity details in the scan data. This paper shows that including those light-based features helps better recognize wood in aerial scans taken from above, which are usually harder to analyze because the data is sparser and from a different viewpoint. Using these features also keeps tree branches more connected and better preserves the tree’s structure for further analysis.
What this means in practice
- •For forest inventory teams: Use improved leaf-wood segmentation with radiometric features to get more accurate wood volume estimates from aerial LiDAR data across different sensors and sites.
- •For environmental monitoring services: Provide better non-destructive tree structure models that maintain branch connectivity for biomass and forest health assessments using data from drones and other aerial scanners.
Authors
Roman Kaharlytskyi, Derek T. Robinson, Roberto Guglielmi
Abstract
Leaf-wood segmentation of individual trees from LiDAR point clouds is essential for quantitative structure models (QSMs) used in non-destructive biomass estimation. Existing segmentation methods typically exclude radiometric features (e.g., intensity, return number) to maximize cross-sensor compatibility. We challenge this design choice by evaluating cross-site and cross-platform generalization: training on the public Heidelberg dataset (terrestrial TLS, 1550nm) and testing on a novel dataset from Ontario, Canada (RPA-LS, 905nm). Results show that geometry-only methods - including state-of-the-art deep learning models trained on high-density LiDAR datasets - fail to generalize to the sparse, top-down geometry of aerial scans, achieving F1 scores <= 0.56. Incorporating radiometric features (intensity, return number, number of returns) improves F1 to 0.61, but more critically, increases wood recall by 119% from 0.16 to 0.35. Furthermore, geometry-only approaches often result in fragmented stem and branch components. We find that leveraging radiometric features preserves greater structural connectivity, resulting in more coherent architectures that are better suited for QSM reconstruction. We demonstrate that while geometric patterns are view-dependent and prone to overfitting scan patterns, radiometric features encode physical material properties that generalize across disparate sensors and environments.