Multispectral LiDAR data for extracting tree points in urban and suburban areas

2025-08-27

AI Summary

This study addresses the challenge of low-accuracy and labor-intensive automatic tree point cloud extraction in complex urban and suburban environments. We propose a deep learning framework that jointly exploits 3D geometric and multispectral LiDAR features. Specifically, we introduce a pseudo-normalized difference vegetation index (P-NDVI), which is fused with raw 3D coordinates for enhanced spectral–geometric representation. Leveraging state-of-the-art point cloud architectures—including Superpoint Transformer and Point Transformer V3—we perform end-to-end semantic segmentation. Experimental results demonstrate a mean intersection-over-union (mIoU) of 85.28%, outperforming a spatial-feature-only baseline by 10.61 percentage points, while maintaining high inference efficiency. The method achieves both robust accuracy and computational scalability, offering a reliable and extensible technical foundation for urban greening planning and intelligent risk assessment of vegetation encroachment near power infrastructure.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Monitoring urban tree dynamics is vital for supporting greening policies and reducing risks to electrical infrastructure. Airborne laser scanning has advanced large-scale tree management, but challenges remain due to complex urban environments and tree variability. Multispectral (MS) light detection and ranging (LiDAR) improves this by capturing both 3D spatial and spectral data, enabling detailed mapping. This study explores tree point extraction using MS-LiDAR and deep learning (DL) models. Three state-of-the-art models are evaluated: Superpoint Transformer (SPT), Point Transformer V3 (PTv3), and Point Transformer V1 (PTv1). Results show the notable time efficiency and accuracy of SPT, with a mean intersection over union (mIoU) of 85.28%. The highest detection accuracy is achieved by incorporating pseudo normalized difference vegetation index (pNDVI) with spatial data, reducing error rate by 10.61 percentage points (pp) compared to using spatial information alone. These findings highlight the potential of MS-LiDAR and DL to improve tree extraction and further tree inventories.

Problem

• Extracting tree points in urban areas using multispectral LiDAR
• Evaluating deep learning models for tree detection accuracy
• Improving tree inventories with spectral and spatial data

Innovation

• Multispectral LiDAR captures 3D spatial and spectral data
• Deep learning models like SPT achieve 85.28% mIoU accuracy
• Pseudo NDVI integration reduces error rate by 10.61%