文献类型：期刊文献

英文题名：SmartQSM: a novel quantitative structure model using sparse-convolution-based point cloud contraction for reconstruction and analysis of individual tree architecture

作者：Yang, Jie[1,2,3] Zhang, Huaiqing[1,2] Li, Jinyang[4,5] Yang, Haoyue[1,2] Gao, Tian[4] Yang, Tingdong[1,2,6] Wang, Jiaxin[3] Zhang, Xiaoli[3] Yun, Ting[7] Duanmu, Yuxin[8] Chen, Sihan[1,2] Shi, Yukai[1,2]

第一作者：Yang, Jie

通信作者：Zhang, HQ[1]

机构：[1]Chinese Acad Forestry, Inst Forest Resource Informat Tech, Beijing 100091, Peoples R China;[2]Natl Forestry & Grassland Sci Data Ctr, Beijing 100091, Peoples R China;[3]Beijing Forestry Univ, Coll Forestry, Beijing 100083, Peoples R China;[4]Chinese Acad Sci, Inst Appl Ecol, Key Lab Forest Ecol & Silviculture, Shenyang 110016, Liaoning, Peoples R China;[5]Shandong Normal Univ, Coll Geog & Environm, Jinan 250358, Shandong, Peoples R China;[6]Beijing Forestry Univ, Sch Technol, Beijing 100083, Peoples R China;[7]Nanjing Forestry Univ, Coll Forestry & Grassland & Soil & Water Conservat, Nanjing 210037, Jiangsu, Peoples R China;[8]Shanghai Univ, Coll Foreign Languages, Shanghai 200444, Peoples R China

年份：2026

卷号：232

起止页码：712-739

外文期刊名：ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING

收录：;EI(收录号：20260519998701);Scopus(收录号：2-s2.0-105028699889);WOS:【SCI-EXPANDED(收录号:WOS:001665835500001)】；

基金：This study was jointly funded by the National Key Research and Development Program of China (Grant number No. 2023YFF1303604) , the Fund of CAS Key Laboratory of Forest Ecology and Silviculture, Institute of Applied Ecology, Chinese Academy of Sciences (KLFES-2042) , the National Natural Science Foundation of China (32171779) , and the Central Public-interest Scientific Institution Basal Research Fund (CAFYBB2022SY035) . We are extremely grateful to Qingda Chen (Institute of Applied Ecologym Chinese Academy of Sciences, a.k.a. IAE) and Yuanyang Hu (IAE) for doing fieldwork; Pengyuan Guan (School of Transportation and Geometric Engineering, Shenyang Jianzhu Univer-sity) for experiment help; Yuxin Zhou (IAE) for image making; Jing-cheng Luo (IAE) , Shuangtian Li (IAE) , Deliang Lu (IAE) and Kexin Lei (Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, a.k.a. IFRIT) for development suggestions; Hao Lei (IFRIT) for content check.

语种：英文

外文关键词：Tree modeling; Tree skeleton extraction; Tree parameter extraction; Deep learning on point clouds; Tree 3D reconstruction; Quantitative structure model

摘要：Tree architecture analysis is fundamental to forestry, but complex trees challenge the accuracy and efficiency of point-cloud-based reconstruction. Here, we present SmartQSM, a novel quantitative structure model designed for reconstructing individual trees and extracting their parameters using ground-based laser scanning data. The method achieves point cloud contraction and forms the thin structures required for skeletonization by iteratively applying a sparse-convolution-based residual U-shaped network (ResUNet) to predict point movement towards the medial axis. This process is integrated with techniques from previous studies to form a complete reconstruction pipeline. Following the organization and QSM-based quantification of 47 individual-scale, 26 organscale, and 8 plot-scale parameters, the proposed method provides comprehensive support for extracting these metrics using the input point cloud and its outputs, including the skeleton and mesh. The performance was verified using the two-period leaf-off LiDAR data of a natural coniferous and broad-leaved mixed forest plot (in Qingyuan county, Liaoning province, China), and 2 open forest datasets. The existing major QSMs was used for comparison. The inference network adopted a three-stage hierarchical spatial compression architecture, initiating with 8 input channels and predicting with multi-layer perceptron. The reconstruction was insensitive to remaining leaves and the model did not have apparent distortion. The processing speed is efficient, about 12,000 points per second. In terms of major architectural parameters, the R2 scores for trunk length, trunk volume, and bole height on the tested two period data of different tree species in the plot reached 0.97, 0.957, and 0.949, respectively, which were 0.043, 0.114, 0.029 higher than existing methods. the R2 scores for branch length, branching angle, and tip deflection angle remained around 0.95. The overestimation of stem volume or aboveground biomass has been alleviated. The high reconstruction quality, efficiency, rich parameters, and unique visual interaction capabilities of the proposed method offer a novel and practical solution for forestry research and broader domains. The implementation code is currently available at: https://github.com/projec t-lightlin/SmartQSM.