Super-resolution of airborne laser scanning point clouds for forest inventory

2026-05-03

AI Summary

This study addresses the limitations of airborne laser scanning (ALS) point clouds—namely their sparsity and noise—which hinder accurate single-tree forest inventory. To overcome this, we propose the 3D Forest Super Resolution (3DFSR) model, a voxel-based U-Net convolutional neural network that achieves cross-platform LiDAR point cloud super-resolution without requiring transfer learning, simultaneously enhancing point density and suppressing noise. Our approach enables ground-based laser scanning–level algorithms to be directly applied to enhanced ALS data across a wide density range of 10–1700 points per square meter. Experimental results demonstrate that the reconstructed point clouds achieve a Chamfer distance of 0.249 m, improve stem detection F1-score from 0.71 to 0.97, reduce diameter-at-breast-height estimation RMSE to 6.43 cm, and yield a trunk volume estimation R² of 0.95.

Abstract

Airborne Laser Scanning (ALS) can collect point clouds across large areas, enabling large-scale forest inventory. However, ALS point clouds are sparse and noisy, resulting in inaccurate individual-tree-level forest inventory, such as stem localization and tree size estimation. To overcome this problem, we propose a deep learning model, 3D Forest Super Resolution (3DFSR), to simultaneously improve point density and reduce noise for ALS forest point cloud. 3DFSR is a voxel-based CNN with a U-Net architecture. The proposed 3DFSR is evaluated on ALS point clouds collected in both temperate forests in the U.S. and boreal forests in Germany. Experimental results demonstrate that 3DFSR can generate finer point clouds of tree structure than other state-of-the-art point cloud super-resolution algorithms, achieving 0.249 m Chamfer Distance and 2.711 m Hausdorff Distance. Furthermore, to verify the effectiveness of 3DFSR point clouds in forest inventory, we conduct stem detection, DBH measurements, and stem reconstruction on both original ALS point clouds and 3DFSR enhanced point clouds. We find that stem detection and reconstruction algorithms developed for TLS/MLS point clouds can directly work on our 3DFSR point clouds, and DBH can be derived with circle-fitting method. F1 score of stem detection is improved from 0.71 on original ALS point clouds to 0.97 on 3DFSR point clouds; DBH estimation improves from 13.45 cm RMSE using allometric equations to 6.43 cm using circle fitting; comparing to stems reconstruction from MLS point clouds, stem reconstructed from 3DFSR point clouds has 0.170 m of Chamfer Distance and 0.377 m of Hausdorff Distance, and 0.95 R2 volume estimation. Finally, we find that the proposed 3DFSR is applicable to process point densities from 10 to 1700 points/m2; it also can be generalized across data collected from different LiDAR platforms without transfer learning.

Problem

• Airborne Laser Scanning
• point cloud super-resolution
• forest inventory
• noise reduction
• point density

Innovation

• point cloud super-resolution
• airborne laser scanning
• forest inventory
• 3D deep learning
• U-Net architecture