@article{ART003374993},
author={Jung in-hye},
title={A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data},
journal={Journal of Internet of Things and Convergence},
issn={2466-0078},
year={2026},
volume={12},
number={4},
pages={21}
TY - JOUR
AU - Jung in-hye
TI - A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data
JO - Journal of Internet of Things and Convergence
PY - 2026
VL - 12
IS - 4
PB - The Korea Internet of Things Society
SP - 21
EP -
SN - 2466-0078
AB - Traditional labor-intensive tree surveys involve substantial time and cost and may produce inconsistent results due to surveyors’ subjective judgments. To overcome these limitations, this study integrates high-resolution 3D mobile LiDAR (Light Detection and Ranging) point cloud data with deep learning to automate urban object classification, individual tree isolation, and species identification. The PointNet++ model was applied to learn hierarchical spatial features directly from point cloud data. Semantic segmentation was first used to distinguish trees from buildings and other artificial structures, followed by instance segmentation to assign a unique ID to each tree. Cross-sectional data at 1.2–1.3 m above ground were then extracted, and stem points were refined using DBSCAN clustering to calculate diameter at breast height and tree height automatically. Experiments conducted along a 6 km corridor in Suwon, Gyeonggi-do, produced a structural attribute database for 1,252 street trees. Validation against field measurements showed that 64.7% of the classified trees had an error rate below 15%. These findings demonstrate the potential of the proposed method to improve forest inventory efficiency and support intelligent urban tree resource management.
KW - LiDAR;Point LiDAR;PointNet++;deep learning;DBH
DO -
UR -
ER -
Jung in-hye. (2026). A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data. Journal of Internet of Things and Convergence, 12(4), 21.
Jung in-hye. 2026, "A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data", Journal of Internet of Things and Convergence, vol.12, no.4 21.
Jung in-hye "A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data" Journal of Internet of Things and Convergence 12.4 21 (2026) : 21.
Jung in-hye. A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data. 2026; 12(4), 21.
Jung in-hye. "A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data" Journal of Internet of Things and Convergence 12, no.4 (2026) : 21.
Jung in-hye. A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data. Journal of Internet of Things and Convergence, 12(4), 21.
Jung in-hye. A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data. Journal of Internet of Things and Convergence. 2026; 12(4) 21.
Jung in-hye. A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data. 2026; 12(4), 21.
Jung in-hye. "A Study on Deep Learning-Based Instance Segmentation and Feature Extraction of Urban Trees Using 3D LiDAR Data" Journal of Internet of Things and Convergence 12, no.4 (2026) : 21.
