Xiaofei Liang

Ms. Xiaofei Liang

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 October 2026 in Land
Verified
Journal Article
Open access
PDF file
YWYongwei Wu
Yongwei Wu
YZYiming Zhang
Yiming Zhang
KSKAI SU
KAI SU
CLChungan Li
Chungan Li
HXHuang Xiaoqian
Huang Xiaoqian
XLXiaofei Liang
Xiaofei Liang
Understanding current forest resource status is fundamental for formulating forestry policies and management plans. 3D laser scanning provides an effective approach for conducting forest resource surveys efficiently and accurately, which is the core background and aim of this study. This research used Unmanned Aerial Vehicle Laser Scanning (UAV-LS) to acquire high-density point cloud data of Eucalyptus(Eucalyptus spp.) and C. lanceolata (Cunninghamia lanceolata) and proposed a refined method by improving the mean-shift algorithm to accurately extract diameter at different heights and individual-tree volume, with measured data adopted for result validation. The improved algorithm effectively filtered dense trunk noise and outperformed K-means, Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and the traditional mean-shift algorithm in detection accuracy. For full-density point cloud data, the mean relative errors of diameter at breast height (DBH) estimation were 2.41% for Eucalyptus and −4.05% for C. lanceolata. The coefficients of determination (R2) of single-tree volume estimation for both tree species were approximately 0.93. Point-cloud density also had a clear effect on estimation accuracy: when the density decreased below 50% of the original density, the accuracy declined sharply, with the maximum absolute error exceeding 86% at 10% density. This study provides technical and theoretical support for improving the efficiency and accuracy of UAV-LS-based forest inventory and offers a reference for high-precision forest assessment under limited data-acquisition conditions.