references.bib

@phdthesis{manickathan_impact_2019,
	title = {Impact of vegetation on urban microclimate},
	copyright = {http://rightsstatements.org/page/InC-NC/1.0/, info:eu-repo/semantics/openAccess},
	url = {http://hdl.handle.net/20.500.11850/379379},
	language = {en},
	urldate = {2024-11-25},
	school = {ETH Zurich},
	author = {Manickathan, Lento},
	collaborator = {{Carmeliet, Jan} and {Blocken, Bert} and {Edwards, Peter}},
	year = {2019},
	doi = {10.3929/ETHZ-B-000379379},
	note = {Artwork Size: 357 p.
Medium: application/pdf
Pages: 357 p.},
	keywords = {Botanical sciences, CFD, Civic \& landscape art, info:eu-repo/classification/ddc/580, info:eu-repo/classification/ddc/710, PIV, Urban microclimate, vegetation, X-ray tomography},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\2IQ9AGFF\\Manickathan - 2019 - Impact of vegetation on urban microclimate.pdf:application/pdf},
}

@misc{majumdar_drag_2024,
	title = {The drag length is key to quantifying tree canopy drag},
	copyright = {arXiv.org perpetual, non-exclusive license},
	url = {https://arxiv.org/abs/2411.01570},
	doi = {10.48550/ARXIV.2411.01570},
	abstract = {The effects of trees on urban flows are often determined using computational fluid dynamics approaches which typically use a quadratic drag formulation based on the leaf-area density \$a\$ and a volumetric drag coefficient \$C\_\{d\}{\textasciicircum}V\$ to model vegetation. In this paper, we develop an analytical model for the flow within a vegetation canopy and identify that the drag length \${\textbackslash}ell\_d = (a C\_d{\textasciicircum}V){\textasciicircum}\{-1\}\$ is the key metric to describe the local tree drag characteristics. A detailed study of the literature suggests that the median \${\textbackslash}ell\_d\$ observed in field experiments is \$21\$ m for trees and \$0.7\$ m for low vegetation (crops). A total of \$168\$ large-eddy simulations are conducted to obtain a closed form of the analytical model. The model allows determining \$a\$ and \$C\_d{\textasciicircum}V\$ from wind-tunnel experiments that typically present the drag characteristics in terms of the classical drag coefficient \$C\_d\$ and the aerodynamic porosity \$α\_L\$. We show that geometric scaling of \${\textbackslash}ell\_d\$ is the appropriate scaling of trees in wind tunnels. Evaluation of \${\textbackslash}ell\_d\$ for numerical simulations and wind-tunnel experiments (assuming geometric scaling \$1:100\$) in literature shows that the median \${\textbackslash}ell\_d\$ in both these cases is about \$5\$ m, suggesting possible overestimation of vegetative drag.},
	urldate = {2024-11-25},
	publisher = {arXiv},
	author = {Majumdar, Dipanjan and Vita, Giulio and Ramponi, Rubina and Glover, Nina and van Reeuwijk, Maarten},
	year = {2024},
	note = {Version Number: 1},
	keywords = {Fluid Dynamics (physics.flu-dyn), FOS: Physical sciences},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\55B8CI4S\\Majumdar et al. - 2024 - The drag length is key to quantifying tree canopy drag.pdf:application/pdf},
}

@article{fu_should_2024,
	title = {Should we care about the level of detail in trees when running urban microscale simulations?},
	volume = {101},
	issn = {22106707},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S2210670723007527},
	doi = {10.1016/j.scs.2023.105143},
	language = {en},
	urldate = {2024-11-25},
	journal = {Sustainable Cities and Society},
	author = {Fu, Runnan and Pađen, Ivan and García-Sánchez, Clara},
	month = feb,
	year = {2024},
	pages = {105143},
	file = {ScienceDirect Full Text PDF:C\:\\Users\\Noah\\Zotero\\storage\\4EJIKR4G\\Fu et al. - 2024 - Should we care about the level of detail in trees when running urban microscale simulations.pdf:application/pdf},
}

@phdthesis{runnan_fu_modeling_2022,
	title = {Modeling tree topology effects on wind},
	language = {en},
	school = {TU Delft},
	author = {{Runnan Fu}},
	year = {2022},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\KKJWYSZM\\_.pdf:application/pdf},
}

@phdthesis{hermann_leaf_2024,
	title = {Leaf it to {AI}: {Mapping} {Urban} {Tree} {Morphology} and {Leaf} {Area} {Index} with {Multimodal} {Deep}-{Learning}},
	language = {en},
	school = {Ecole Polytechnique Federale de Lausanne},
	author = {Hermann, Theo},
	year = {2024},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\MVETQVKP\\Hermann - Leaf it to AI Mapping Urban Tree Morphology and Leaf Area Index with Multimodal Deep-Learning.pdf:application/pdf},
}

@misc{xiao_individual_2021,
	title = {Individual {Tree} {Detection} and {Crown} {Delineation} with {3D} {Information} from {Multi}-view {Satellite} {Images}},
	copyright = {arXiv.org perpetual, non-exclusive license},
	url = {https://arxiv.org/abs/2107.00592},
	doi = {10.48550/ARXIV.2107.00592},
	abstract = {Individual tree detection and crown delineation (ITDD) are critical in forest inventory management and remote sensing based forest surveys are largely carried out through satellite images. However, most of these surveys only use 2D spectral information which normally has not enough clues for ITDD. To fully explore the satellite images, we propose a ITDD method using the orthophoto and digital surface model (DSM) derived from the multi-view satellite data. Our algorithm utilizes the top-hat morphological operation to efficiently extract the local maxima from DSM as treetops, and then feed them to a modi-fied superpixel segmentation that combines both 2D and 3D information for tree crown delineation. In subsequent steps, our method incorporates the biological characteristics of the crowns through plant allometric equation to falsify potential outliers. Experiments against manually marked tree plots on three representative regions have demonstrated promising results - the best overall detection accuracy can be 89\%.},
	urldate = {2024-11-18},
	publisher = {arXiv},
	author = {Xiao, Changlin and Qin, Rongjun and Xie, Xiao and Huang, Xu},
	year = {2021},
	note = {Version Number: 1},
	keywords = {Computer Vision and Pattern Recognition (cs.CV), FOS: Computer and information sciences},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\PRTVZG8W\\Xiao et al. - 2021 - Individual Tree Detection and Crown Delineation with 3D Information from Multi-view Satellite Images.pdf:application/pdf},
}

@phdthesis{lessie_m_ortega-cordova_urban_2018,
	title = {Urban {Vegetation} {Modeling} {3D} {Levels} of {Detail}},
	language = {en},
	school = {TU Delft},
	author = {{Lessie M Ortega-Córdova}},
	year = {2018},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\VT2J3637\\Ortega-Córdova - Urban Vegetation Modeling 3D Levels of Detail.pdf:application/pdf},
}

@phdthesis{geert_jan_de_groot_automatic_2020,
	title = {Automatic construction of {3D} tree models in multiple levels of detail from airborne {LiDAR} data},
	language = {en},
	school = {TU Delft},
	author = {{Geert Jan de Groot}},
	year = {2020},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\C7CI23ZW\\Jan - Automatic construction of 3D tree models in multiple levels of detail from airborne LiDAR data.pdf:application/pdf},
}

@article{buccolieri_review_2018,
	title = {Review on urban tree modelling in {CFD} simulations: {Aerodynamic}, deposition and thermal effects},
	volume = {31},
	issn = {1618-8667},
	shorttitle = {Review on urban tree modelling in {CFD} simulations},
	url = {https://www.sciencedirect.com/science/article/pii/S1618866717304600},
	doi = {10.1016/j.ufug.2018.03.003},
	abstract = {This paper reviews current parameterizations developed and implemented within Computational Fluid Dynamics models for the study of the effects linking vegetation, mainly trees, to urban air quality and thermal conditions. In the literature, passive mitigation via deposition is parametrized as a volumetric sink term in the transport equation of pollutants, while a volumetric source term is used for particle resuspension. The aerodynamics effects are modelled via source and sink terms of momentum, turbulent kinetic energy and turbulent dissipation rate. A volumetric cooling power is finally considered to account for the thermal (transpirational cooling) effects of vegetation. The most recent applications are also summarized with a focus on the relative importance of both aerodynamic and deposition effects, together with recent studies evaluating thermal effects. Those studies have shown that the aerodynamic effects of trees are stronger than the positive effects of deposition, however locally the pollutant concentration increases or decreases depending on the complex inter-relation between local factors such as vegetation type and density, meteorological conditions, street geometry, pollutant characteristics and emission rates. Unlike aerodynamic and deposition effects on pollutant dispersion which were also found in street far from trees, the thermal effects were in general locally restricted to the close vicinity of the vegetation and to the street canyon itself. Future requirements in CFD modelling include more in depth investigation of resuspension and thermal effects, as well as of the VOCs emissions and chemical reactions. The overall objective of this review is to provide the scientific community with a comprehensive summary on the current parameterizations of urban vegetation in CFD modelling and constitutes the starting point for the development of new parametrizations in CFD as well as in mesoscale models.},
	urldate = {2024-11-18},
	journal = {Urban Forestry \& Urban Greening},
	author = {Buccolieri, Riccardo and Santiago, Jose-Luis and Rivas, Esther and Sanchez, Beatriz},
	month = apr,
	year = {2018},
	keywords = {Aerodynamic, CFD modelling, Deposition and thermal effects, Trees, Urban air quality, Vegetation parameterizations},
	pages = {212--220},
	file = {ScienceDirect Full Text PDF:C\:\\Users\\Noah\\Zotero\\storage\\ZJUCBDSZ\\Buccolieri et al. - 2018 - Review on urban tree modelling in CFD simulations Aerodynamic, deposition and thermal effects.pdf:application/pdf;ScienceDirect Snapshot:C\:\\Users\\Noah\\Zotero\\storage\\38SQFRM3\\S1618866717304600.html:text/html},
}

@article{kamoske_leaf_2019,
	title = {Leaf area density from airborne {LiDAR}: {Comparing} sensors and resolutions in a temperate broadleaf forest ecosystem},
	volume = {433},
	issn = {03781127},
	shorttitle = {Leaf area density from airborne {LiDAR}},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0378112718315561},
	doi = {10.1016/j.foreco.2018.11.017},
	language = {en},
	urldate = {2024-11-18},
	journal = {Forest Ecology and Management},
	author = {Kamoske, Aaron G. and Dahlin, Kyla M. and Stark, Scott C. and Serbin, Shawn P.},
	month = feb,
	year = {2019},
	pages = {364--375},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\9EAK29EY\\Kamoske et al. - 2019 - Leaf area density from airborne LiDAR Comparing sensors and resolutions in a temperate broadleaf fo.pdf:application/pdf},
}

@article{groot_predicting_2011,
	title = {Predicting maximum branch diameter from crown dimensions, stand characteristics and tree species},
	volume = {87},
	issn = {0015-7546, 1499-9315},
	url = {http://pubs.cif-ifc.org/doi/10.5558/tfc2011-053},
	doi = {10.5558/tfc2011-053},
	abstract = {Forest resource inventories must include wood quality information to support the optimum use of wood fibre. The objective of this study was to develop models relating maximum live branch diameter (MBD), which affects lumber value, to tree and stand characteristics that can be measured through current and emerging remote sensing technologies. Using non-linear mixed effects models for six Canadian conifer species, as well as for three broad-leaved species, MBD was related to crown radius, tree height, crown length, stand basal area, and basal area of trees larger than the subject tree. Models that included only individual tree characteristics (crown radius, tree height, and crown length) did not perform as well as models that additionally included stand characteristics (stand basal area and basal area of larger trees). Models that took into account tree species performed better than models that did not; in particular, broadleaved species had much thicker branches than conifers. The best model did not show bias with respect to independent variables and had root mean square error of 0.32 cm. For the best model, prediction error was not related to silvicultural treatment. These model characteristics strongly support the potential to successfully predict MBD from remotely sensed data.},
	language = {en},
	number = {04},
	urldate = {2024-11-18},
	journal = {The Forestry Chronicle},
	author = {Groot, Arthur and Schneider, Robert},
	month = aug,
	year = {2011},
	pages = {542--551},
	file = {Full Text:C\:\\Users\\Noah\\Zotero\\storage\\ZLLM3FMC\\Groot and Schneider - 2011 - Predicting maximum branch diameter from crown dimensions, stand characteristics and tree species.pdf:application/pdf},
}

@article{kastner_eddy3d_2022,
	title = {{Eddy3D}: {A} toolkit for decoupled outdoor thermal comfort simulations in urban areas},
	volume = {212},
	issn = {03601323},
	shorttitle = {{Eddy3D}},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0360132321010301},
	doi = {10.1016/j.buildenv.2021.108639},
	language = {en},
	urldate = {2024-11-18},
	journal = {Building and Environment},
	author = {Kastner, Patrick and Dogan, Timur},
	month = mar,
	year = {2022},
	pages = {108639},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\ZEZIRHYK\\Kastner and Dogan - 2022 - Eddy3D A toolkit for decoupled outdoor thermal comfort simulations in urban areas.pdf:application/pdf},
}

@article{bartkowicz_morphological_2023,
	title = {Morphological plasticity of six tree species with different light demands growing in multi-layered deciduous forests in {Central} {Europe}},
	volume = {142},
	issn = {1612-4677},
	url = {https://doi.org/10.1007/s10342-023-01584-7},
	doi = {10.1007/s10342-023-01584-7},
	abstract = {Tree allometry is a plastic feature, and scaling parameters can vary considerably depending on phylogeny, life strategies, growth conditions and ontogeny. We hypothesized that in multi-layered forests growing on rich sites and driven by stand dynamics without stand-replacing disturbances, light is a primary driver of allometric relationships and that the morphological plasticity of tree species is closely associated with their shade tolerance. We quantified and compared the morphological properties of six species that form a shade tolerance gradient: Alnus glutinosa (L.) Gaertner, Quercus robur L., Fraxinus excelsior L., Ulmus laevis Pall., Tilia cordata Miller and Carpinus betulus L. The relationships between tree height and local stand density as predictors and dbh, crown width, crown length and crown volume as response variables were characterized. We found that in the lower stand layer the values of crown parameters increased with tree height at a lower rate in light-adapted than in shade-tolerant species. Conversely, the response of morphological traits on competition was stronger in light-adapted species than in shade-tolerant species. The ratio of crown width-to-crown length was not associated with light demand. Apart from ash, which demonstrated a different allocation pattern, between-species differences in the slenderness ratio were insignificant. Allometry and sensitivity to competition varied in trees growing in the upper and lower stand layers. Our results indicate that the dichotomy of basic growth strategies of stress tolerance versus stress avoidance is overly simplistic and fails to consider social status and species-specific features such as apical control.},
	language = {en},
	number = {5},
	urldate = {2024-11-18},
	journal = {European Journal of Forest Research},
	author = {Bartkowicz, Leszek and Paluch, Jarosław},
	month = oct,
	year = {2023},
	keywords = {Adaptive forestry, Crown morphology, Life strategy, Shade tolerance, Slenderness ratio, Tree architecture},
	pages = {1177--1195},
	file = {Full Text PDF:C\:\\Users\\Noah\\Zotero\\storage\\FNAQULKB\\Bartkowicz and Paluch - 2023 - Morphological plasticity of six tree species with different light demands growing in multi-layered d.pdf:application/pdf},
}

@inproceedings{hoetzlein_procedural_2022,
	address = {New York, NY, USA},
	series = {{FDG} '22},
	title = {A {Procedural} {Model} for {Diverse} {Tree} {Species}},
	isbn = {978-1-4503-9795-7},
	url = {https://doi.org/10.1145/3555858.3564251},
	doi = {10.1145/3555858.3564251},
	abstract = {The modeling of trees represents a unique and classical challenge in computer graphics. Models of 3D trees must express the form, complexity, structure, growth and diversity of real trees. Presently the most common methods for the modeling of 3D trees include a) user-based creative modeling, b) direct geometric capture such as LIDAR and photogrammetry, or c) indirect methods such as machine learning from images. These techniques often require significant human effort, large amounts of data, considerable computation resources, or any of the above. While there are methods that consider the direct procedural generation of trees, current models often require some human supervision to focus on naturally plausible variants. Instead, our approach is to construct a botanically-inspired, harmonic, procedural model for trees which directly produces realistic yet diverse trees.},
	urldate = {2024-11-18},
	booktitle = {Proceedings of the 17th {International} {Conference} on the {Foundations} of {Digital} {Games}},
	publisher = {Association for Computing Machinery},
	author = {Hoetzlein, Rama},
	month = nov,
	year = {2022},
	pages = {1--8},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\S4AAD79X\\Hoetzlein - 2022 - A Procedural Model for Diverse Tree Species.pdf:application/pdf},
}

@misc{lee_graph_2017,
	title = {A graph cut approach to {3D} tree delineation, using integrated airborne {LiDAR} and hyperspectral imagery},
	url = {http://arxiv.org/abs/1701.06715},
	abstract = {Recognising individual trees within remotely sensed imagery has important applications in forest ecology and management. Several algorithms for tree delineation have been suggested, mostly based on locating local maxima or inverted basins in raster canopy height models (CHMs) derived from Light Detection And Ranging (LiDAR) data or photographs. However, these algorithms often lead to inaccurate estimates of forest stand characteristics due to the limited information content of raster CHMs. Here we develop a 3D tree delineation method which uses graph cut to delineate trees from the full 3D LiDAR point cloud, and also makes use of any optical imagery available (hyperspectral imagery in our case). First, conventional methods are used to locate local maxima in the CHM and generate an initial map of trees. Second, a graph is built from the LiDAR point cloud, fused with the hyperspectral data. For computational efficiency, the feature space of hyperspectral imagery is reduced using robust PCA. Third, a multi-class normalised cut is applied to the graph, using the initial map of trees to constrain the number of clusters and their locations. Finally, recursive normalised cut is used to subdivide, if necessary, each of the clusters identified by the initial analysis. We call this approach Multiclass Cut followed by Recursive Cut (MCRC). The effectiveness of MCRC was tested using three datasets: i) NewFor, ii) a coniferous forest in the Italian Alps, and iii) a deciduous woodland in the UK. The performance of MCRC was usually superior to that of other delineation methods, and was further improved by including high-resolution optical imagery. Since MCRC delineates the entire LiDAR point cloud in 3D, it allows individual crown characteristics to be measured. By making full use of the data available, graph cut has the potential to considerably improve the accuracy of tree delineation.},
	urldate = {2024-11-18},
	publisher = {arXiv},
	author = {Lee, Juheon and Coomes, David and Schonlieb, Carola-Bibiane and Cai, Xiaohao and Lellmann, Jan and Dalponte, Michele and Malhi, Yadvinder and Butt, Nathalie and Morecroft, Mike},
	month = jan,
	year = {2017},
	note = {arXiv:1701.06715 [cs]},
	keywords = {Computer Science - Computer Vision and Pattern Recognition, To read},
	file = {Preprint PDF:C\:\\Users\\Noah\\Zotero\\storage\\CUMV2ZEP\\Lee et al. - 2017 - A graph cut approach to 3D tree delineation, using integrated airborne LiDAR and hyperspectral image.pdf:application/pdf;Snapshot:C\:\\Users\\Noah\\Zotero\\storage\\PPXQ4BD6\\1701.html:text/html},
}

@article{chen_tree_2023,
	title = {Tree {Species} {Classification} in {Subtropical} {Natural} {Forests} {Using} {High}-{Resolution} {UAV} {RGB} and {SuperView}-1 {Multispectral} {Imageries} {Based} on {Deep} {Learning} {Network} {Approaches}: {A} {Case} {Study} within the {Baima} {Snow} {Mountain} {National} {Nature} {Reserve}, {China}},
	volume = {15},
	copyright = {https://creativecommons.org/licenses/by/4.0/},
	issn = {2072-4292},
	shorttitle = {Tree {Species} {Classification} in {Subtropical} {Natural} {Forests} {Using} {High}-{Resolution} {UAV} {RGB} and {SuperView}-1 {Multispectral} {Imageries} {Based} on {Deep} {Learning} {Network} {Approaches}},
	url = {https://www.mdpi.com/2072-4292/15/10/2697},
	doi = {10.3390/rs15102697},
	abstract = {Accurate information on dominant tree species and their spatial distribution in subtropical natural forests are key ecological monitoring factors for accurately characterizing forest biodiversity, depicting the tree competition mechanism and quantitatively evaluating forest ecosystem stability. In this study, the subtropical natural forest in northwest Yunnan province of China was selected as the study area. Firstly, an object-oriented multi-resolution segmentation (MRS) algorithm was used to segment individual tree crowns from the UAV RGB imagery and satellite multispectral imagery in the forests with different densities (low (547 n/ha), middle (753 n/ha) and high (1040 n/ha)), and parameters of the MRS algorithm were tested and optimized for accurately extracting the tree crown and position information of the individual tree. Secondly, the texture metrics of the UAV RGB imagery and the spectral metrics of the satellite multispectral imagery within the individual tree crown were extracted, and the random forest algorithm and three deep learning networks constructed in this study were utilized to classify the five dominant tree species. Finally, we compared and evaluated the performance of the random forest algorithm and three deep learning networks for dominant tree species classification using the field measurement data, and the influence of the number of training samples on the accuracy of dominant tree species classification using deep learning networks was investigated. The results showed that: (1) Stand density had little influence on individual tree segmentation using the object-oriented MRS algorithm. In the forests with different stand densities, the F1 score of individual tree segmentation based on satellite multispectral imagery was 71.3–74.7\%, and that based on UAV high-resolution RGB imagery was 75.4–79.2\%. (2) The overall accuracy of dominant tree species classification using the light-weight network MobileNetV2 (OA = 71.11–82.22\%), residual network ResNet34 (OA = 78.89–91.11\%) and dense network DenseNet121 (OA = 81.11–94.44\%) was higher than that of the random forest algorithm (OA = 60.00–64.44\%), among which DenseNet121 had the highest overall accuracy. Texture metrics improved the overall accuracy of dominant tree species classification. (3) For the three deep learning networks, the changes in overall accuracy of dominant tree species classification influenced by the number of training samples were 2.69–4.28\%.},
	language = {en},
	number = {10},
	urldate = {2024-11-18},
	journal = {Remote Sensing},
	author = {Chen, Xianggang and Shen, Xin and Cao, Lin},
	month = may,
	year = {2023},
	pages = {2697},
	file = {Full Text:C\:\\Users\\Noah\\Zotero\\storage\\LWJG2D5T\\Chen et al. - 2023 - Tree Species Classification in Subtropical Natural Forests Using High-Resolution UAV RGB and SuperVi.pdf:application/pdf},
}

@misc{racine_tree_2021,
	title = {Tree species, crown cover, and age as determinants of the vertical distribution of airborne {LiDAR} returns},
	copyright = {Creative Commons Attribution 4.0 International},
	url = {https://arxiv.org/abs/2104.05057},
	doi = {10.48550/ARXIV.2104.05057},
	abstract = {Light detection and ranging (LiDAR) provides information on the vertical structure of forest stands enabling detailed and extensive ecosystem study. The vertical structure is often summarized by scalar features and data-reduction techniques that limit the interpretation of results. Instead, we quantified the influence of three variables, species, crown cover, and age, on the vertical distribution of airborne LiDAR returns from forest stands. We studied 5,428 regular, even-aged stands in Quebec (Canada) with five dominant species: balsam fir (Abies balsamea (L.) Mill.), paper birch (Betula papyrifera Marsh), black spruce (Picea mariana (Mill.) BSP), white spruce (Picea glauca Moench) and aspen (Populus tremuloides Michx.). We modeled the vertical distribution against the three variables using a functional general linear model and a novel nonparametric graphical test of significance. Results indicate that LiDAR returns from aspen stands had the most uniform vertical distribution. Balsam fir and white birch distributions were similar and centered at around 50\% of the stand height, and black spruce and white spruce distributions were skewed to below 30\% of stand height (p\<0.001). Increased crown cover concentrated the distributions around 50\% of stand height. Increasing age gradually shifted the distributions higher in the stand for stands younger than 70-years, before plateauing and slowly declining at 90-120 years. Results suggest that the vertical distributions of LiDAR returns depend on the three variables studied.},
	urldate = {2024-11-18},
	publisher = {arXiv},
	author = {Racine, Etienne and Coops, Nicholas C. and Bégin, Jean and Myllymäki, Mari},
	year = {2021},
	note = {Version Number: 2},
	keywords = {FOS: Biological sciences, Quantitative Methods (q-bio.QM)},
	file = {PDF:C\:\\Users\\Noah\\Zotero\\storage\\MG4KXJTU\\Racine et al. - 2021 - Tree species, crown cover, and age as determinants of the vertical distribution of airborne LiDAR re.pdf:application/pdf},
}

@article{mao_dbh_2023,
	title = {{DBH} {Estimation} for {Individual} {Tree}: {Two}-{Dimensional} {Images} or {Three}-{Dimensional} {Point} {Clouds}?},
	volume = {15},
	copyright = {https://creativecommons.org/licenses/by/4.0/},
	issn = {2072-4292},
	shorttitle = {{DBH} {Estimation} for {Individual} {Tree}},
	url = {https://www.mdpi.com/2072-4292/15/16/4116},
	doi = {10.3390/rs15164116},
	abstract = {Accurate forest parameters are crucial for ecological protection, forest resource management and sustainable development. The rapid development of remote sensing can retrieve parameters such as the leaf area index, cluster index, diameter at breast height (DBH) and tree height at different scales (e.g., plots and stands). Although some LiDAR satellites such as GEDI and ICESAT-2 can measure the average tree height in a certain area, there is still a lack of effective means for obtaining individual tree parameters using high-resolution satellite data, especially DBH. The objective of this study is to explore the capability of 2D image-based features (texture and spectrum) in estimating the DBH of individual tree. Firstly, we acquired unmanned aerial vehicle (UAV) LiDAR point cloud data and UAV RGB imagery, from which digital aerial photography (DAP) point cloud data were generated using the structure-from-motion (SfM) method. Next, we performed individual tree segmentation and extracted the individual tree crown boundaries using the DAP and LiDAR point cloud data, respectively. Subsequently, the eight 2D image-based textural and spectral metrics and 3D point-cloud-based metrics (tree height and crown diameters) were extracted from the tree crown boundaries of each tree. Then, the correlation coefficients between each metric and the reference DBH were calculated. Finally, the capabilities of these metrics and different models, including multiple linear regression (MLR), random forest (RF) and support vector machine (SVM), in the DBH estimation were quantitatively evaluated and compared. The results showed that: (1) The 2D image-based textural metrics had the strongest correlation with the DBH. Among them, the highest correlation coefficient of −0.582 was observed between dissimilarity, variance and DBH. When using textural metrics alone, the estimated DBH accuracy was the highest, with a RMSE of only 0.032 and RMSE\% of 16.879\% using the MLR model; (2) Simply feeding multi-features, such as textural, spectral and structural metrics, into the machine learning models could not have led to optimal results in individual tree DBH estimations; on the contrary, it could even reduce the accuracy. In general, this study indicated that the 2D image-based textural metrics have great potential in individual tree DBH estimations, which could help improve the capability to efficiently and meticulously monitor and manage forests on a large scale.},
	language = {en},
	number = {16},
	urldate = {2024-11-18},
	journal = {Remote Sensing},
	author = {Mao, Zhihui and Lu, Zhuo and Wu, Yanjie and Deng, Lei},
	month = aug,
	year = {2023},
	pages = {4116},
	file = {Full Text:C\:\\Users\\Noah\\Zotero\\storage\\LCKCH9FL\\Mao et al. - 2023 - DBH Estimation for Individual Tree Two-Dimensional Images or Three-Dimensional Point Clouds.pdf:application/pdf},
}

@techreport{bry_tree_2024,
	title = {Tree object detection using airborne images and {LiDAR} point clouds},
	url = {https://zokszy.github.io/Geodan-internship-report/},
	urldate = {2024-12-02},
	author = {Bry, Alexandre},
	year = {2024},
	file = {Internship Report:C\:\\Users\\Noah\\Zotero\\storage\\CQMWHVFQ\\Geodan-internship-report.html:text/html;PDF:C\:\\Users\\Noah\\Zotero\\storage\\Y9YZG4BU\\Bry - 2024 - Tree object detection using airborne images and LiDAR point clouds.pdf:application/pdf},
}

@article{weinstein_deepforest_2020,
	title = {{DeepForest}: {A} {\textless}span style="font-variant:small-caps;"{\textgreater}{Python}{\textless}/span{\textgreater} package for {RGB} deep learning tree crown delineation},
	volume = {11},
	issn = {2041-210X, 2041-210X},
	shorttitle = {{DeepForest}},
	url = {https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.13472},
	doi = {10.1111/2041-210X.13472},
	abstract = {Abstract
            
              
                
                  
                    Remote sensing of forested landscapes can transform the speed, scale and cost of forest research. The delineation of individual trees in remote sensing images is an essential task in forest analysis. Here we introduce a new
                    Python
                    package, DeepForest that detects individual trees in high resolution RGB imagery using deep learning.
                  
                
                
                  While deep learning has proven highly effective in a range of computer vision tasks, it requires large amounts of training data that are typically difficult to obtain in ecological studies. DeepForest overcomes this limitation by including a model pretrained on over 30 million algorithmically generated crowns from 22 forests and fine‐tuned using 10,000 hand‐labelled crowns from six forests.
                
                
                  The package supports the application of this general model to new data, fine tuning the model to new datasets with user labelled crowns, training new models and evaluating model predictions. This simplifies the process of using and retraining deep learning models for a range of forests, sensors and spatial resolutions.
                
                
                  We illustrate the workflow of DeepForest using data from the National Ecological Observatory Network, a tropical forest in French Guiana, and street trees from Portland, Oregon.},
	language = {en},
	number = {12},
	urldate = {2024-12-02},
	journal = {Methods in Ecology and Evolution},
	author = {Weinstein, Ben G. and Marconi, Sergio and Aubry‐Kientz, Mélaine and Vincent, Gregoire and Senyondo, Henry and White, Ethan P.},
	editor = {Record, Sydne},
	month = dec,
	year = {2020},
	pages = {1743--1751},
	file = {Full Text:C\:\\Users\\Noah\\Zotero\\storage\\NN5ED5VS\\Weinstein et al. - 2020 - DeepForest A Python package for RGB deep learning tre.pdf:application/pdf},
}