Quantitative Analysis of Superior Structural Features in Hickory Trees Based on Terrestrial LiDAR Point Cloud and Machine Learning

被引:0
作者
Chen, Yi [1 ]
Yang, Yinhui [1 ]
Xu, Zhuangzhi [2 ]
Ding, Lizhong [3 ]
Wang, Weiyu [3 ]
Huang, Jianqin [4 ]
机构
[1] Zhejiang A&F Univ, Sch Math & Comp Sci, Hangzhou 311300, Peoples R China
[2] East China Acad Inventory & Planning NFGA, Hangzhou 310019, Peoples R China
[3] Hangzhou Linan Dist Agr & Forestry Technol Extens, Hangzhou 311300, Peoples R China
[4] Zhejiang A&F Univ, State Key Lab Subtrop Silviculture, Hangzhou 311300, Peoples R China
关键词
3D digitization; quantitative structural model; machine learning; ground-based lidar;
D O I
10.3390/f16060878
中图分类号
S7 [林业];
学科分类号
0829 ; 0907 ;
摘要
The structural characteristics of hickory trees exhibit a significant correlation with their fruit yield. As a distinctive high-quality nut of Zhejiang Province, hickory is a unique high-end dry fruit and woody oil plant in China. However, the long growth cycle and extended maturation period make their management particularly challenging, especially in the absence of high-precision 3D digital models. This study aims to optimize hickory tree management and identify trees with the most optimal structural features. It employs gradient-boosted machine learning modeling based on 23 key tree characteristics, transforming the experiential knowledge of forest farmers into quantifiable parameters. The consensus model achieved an LOOCV average accuracy of 87%, a training set accuracy of 100%, and a test set accuracy of 78%. Through this approach, three structural parameters that significantly impact the hickory tree were identified: the number of branches, the total length of all branches, and the crown base height from the ground. These parameters were used to select trees with superior structural traits. Furthermore, a novel method based on distance metrics was developed to assess the structural similarity of trees. This research not only highlights the importance of incorporating tree structural characteristics into forest management practices but also demonstrates how modern technological tools can enhance the productivity and economic returns of hickory forests. Through this integration, both the sustainability and economic viability of hickory forests are improved.
引用
收藏
页数:28
相关论文
共 50 条
[1]   Time and amount of supplemental irrigation at different distances from tree trunks influence on morphological characteristics and physiological responses of rainfed fig trees under drought conditions [J].
Abdolahipour, Mohammad ;
Kamgar-Haghighi, Ali Akbar ;
Sepaskhah, Ali Reza ;
Zand-Parsa, Shahrokh ;
Honar, Tooraj ;
Razzaghi, Fatemeh .
SCIENTIA HORTICULTURAE, 2019, 253 :241-254
[2]   Terrestrial Laser Scanning (TLS) for tree structure studies: a review of methods for wood-leaf classifications from 3D point clouds [J].
Arrizza, S. ;
Marras, S. ;
Ferrara, R. ;
Pellizzaro, G. .
REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT, 2024, 36
[3]   Semi-direct tree reconstruction using terrestrial LiDAR point cloud data [J].
Bailey, Brian N. ;
Ochoa, Miguel H. .
REMOTE SENSING OF ENVIRONMENT, 2018, 208 :133-144
[4]  
Berger YG, 2020, J OFF STAT, V36, P237, DOI 10.2478/JOS-2020-0013
[5]   Voxel-Based Automatic Tree Detection and Parameter Retrieval from Terrestrial Laser Scans for Plot-Wise Forest Inventory [J].
Brolly, Gabor ;
Kiraly, Geza ;
Lehtomaki, Matti ;
Liang, Xinlian .
REMOTE SENSING, 2021, 13 (04) :1-23
[6]   Nondestructive estimates of above-ground biomass using terrestrial laser scanning [J].
Calders, Kim ;
Newnham, Glenn ;
Burt, Andrew ;
Murphy, Simon ;
Raumonen, Pasi ;
Herold, Martin ;
Culvenor, Darius ;
Avitabile, Valerio ;
Disney, Mathias ;
Armston, John ;
Kaasalainen, Mikko .
METHODS IN ECOLOGY AND EVOLUTION, 2015, 6 (02) :198-208
[7]   An Individual Tree Segmentation Method That Combines LiDAR Data and Spectral Imagery [J].
Chen, Xingwang ;
Wang, Ruirui ;
Shi, Wei ;
Li, Xiuting ;
Zhu, Xianhao ;
Wang, Xiaoyan .
FORESTS, 2023, 14 (05)
[8]   Tree architecture, light interception and water-use related traits are controlled by different genomic regions in an apple tree core collection [J].
Coupel-Ledru, Aude ;
Pallas, Benoit ;
Delalande, Magalie ;
Segura, Vincent ;
Guitton, Baptiste ;
Muranty, Helene ;
Durel, Charles-Eric ;
Regnard, Jean-Luc ;
Costes, Evelyne .
NEW PHYTOLOGIST, 2022, 234 (01) :209-226
[9]   Simulating an Autonomously Operating Low-Cost Static Terrestrial LiDAR for Multitemporal Maize Crop Height Measurements [J].
Crommelinck, Sophie ;
Hoefle, Bernhard .
REMOTE SENSING, 2016, 8 (03)
[10]   MDC-Net: a multi-directional constrained and prior assisted neural network for wood and leaf separation from terrestrial laser scanning [J].
Dai, Wenxia ;
Jiang, Yiheng ;
Zeng, Wen ;
Chen, Ruibo ;
Xu, Yongyang ;
Zhu, Ningning ;
Xiao, Wen ;
Dong, Zhen ;
Guan, Qingfeng .
INTERNATIONAL JOURNAL OF DIGITAL EARTH, 2023, 16 (01) :1224-1245