No Cover Image

Journal article 20 views 2 downloads

SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests

Diego Laino, Carlos Cabo, Celestino Ordóñez, Rodolfo Bolanos, Romain Janvier, Federico Giulioni, Miriam Herrmann, Andrew Hudak, Russell Parsons, Cristina Santin

Forestry: An International Journal of Forest Research, Volume: 99, Issue: 2

  • 71694.VoR.pdf

    PDF | Version of Record

    ©The Author(s) 2025. This is an Open Access article distributed under the terms of the Creative Commons Attribution License.

    Download (3.07MB)

Check full text

DOI (Published version): 10.1093/forestry/cpaf062

Abstract

Semantic segmentation of point clouds using deep learning (DL) has been the subject of research in forestry in recent years due to its potential applications. Several scientific and management disciplines, such as biodiversity monitoring, ecosystem carbon assessments, or forest management could bene...

Full description

Published in: Forestry: An International Journal of Forest Research
ISSN: 0015-752X 1464-3626
Published: Oxford University Press (OUP) 2026
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa71694
first_indexed 2026-04-01T09:32:14Z
last_indexed 2026-04-03T03:30:52Z
id cronfa71694
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2026-04-01T10:36:31.2682553</datestamp><bib-version>v2</bib-version><id>71694</id><entry>2026-04-01</entry><title>SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests</title><swanseaauthors/><date>2026-04-01</date><abstract>Semantic segmentation of point clouds using deep learning (DL) has been the subject of research in forestry in recent years due to its potential applications. Several scientific and management disciplines, such as biodiversity monitoring, ecosystem carbon assessments, or forest management could benefit from this technique. However, it requires manual segmentation of point clouds to be used as training data. This process is highly labour-intensive and time-consuming, and there is a notable lack of publicly available datasets to support the development of accurate DL semantic segmentation models for forestry and forest ecology applications. Here, we present SegmentedForests, a curated dataset of manually segmented ground-based point clouds from forest plots, specifically designed to facilitate the training and validation of semantic segmentation models. This publicly available dataset contains &gt;920 million labelled points from 14 forest plots, acquired using both terrestrial laser scanning (TLS) and mobile laser scanning (MLS) technologies. It covers two hectares of broadleaf, conifer, and mixed stands from different bioclimatic regions and features &gt;1600 trees across 16 tree species. Each point cloud is labelled into multiple vegetation classes (up to 16), such as tree stems, branches, grass, shrubs, and down wood, as well as non-vegetation elements commonly present in forest scenes, including rocks, people, and stakes. Data splits to facilitate DL model development using our dataset are provided as well. The dataset is available at https://zenodo.org/records/17396681. By releasing this annotated dataset, we seek to address the critical need for publicly available, high-quality training data for DL models that perform semantic segmentation of ground-based point clouds in forest ecosystems.</abstract><type>Journal Article</type><journal>Forestry: An International Journal of Forest Research</journal><volume>99</volume><journalNumber>2</journalNumber><paginationStart/><paginationEnd/><publisher>Oxford University Press (OUP)</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0015-752X</issnPrint><issnElectronic>1464-3626</issnElectronic><keywords>forestry; deep learning; artificial intelligence; TLS; supervised segmentation; 3D computer vision</keywords><publishedDay>1</publishedDay><publishedMonth>4</publishedMonth><publishedYear>2026</publishedYear><publishedDate>2026-04-01</publishedDate><doi>10.1093/forestry/cpaf062</doi><url/><notes>A correction has been published: Forestry: An International Journal of Forest Research, Volume 99, Issue 2, April 2026, cpag018, https://doi.org/10.1093/forestry/cpag018</notes><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm/><funders>This work was supported by the UK NERC project(NE/T001194/1): &#x2018;Advancing 3D Fuel Mapping for Wildfire Behaviour and Risk Mitigation Modelling&#x2019;, the Spanish Knowledge Generation project (PID2021-126790NB-I00): &#x2018;Advancing carbon emission estimations from wildfires applying artificial intelligence to 3D terrestrial point clouds&#x2019;, (grant PRE2022-104159) funded by MCIN/AEI/10.13039/501100011033 and FSE+, research grant &#x2018;FIREPROs&#x2019; (IDE/2024/000780) funded by the Principality of Asturias Government (Spain), COST Action 3DForEcoTech CA20118 supported by COST (European Cooperation in Science and Technology) and the US DoD SERDP/ESTCP projects (RC201025), (RC23-7626), and (RC20-1046).</funders><projectreference/><lastEdited>2026-04-01T10:36:31.2682553</lastEdited><Created>2026-04-01T10:26:11.3754926</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Biosciences, Geography and Physics - Geography</level></path><authors><author><firstname>Diego</firstname><surname>Laino</surname><order>1</order></author><author><firstname>Carlos</firstname><surname>Cabo</surname><order>2</order></author><author><firstname>Celestino</firstname><surname>Ord&#xF3;&#xF1;ez</surname><order>3</order></author><author><firstname>Rodolfo</firstname><surname>Bolanos</surname><order>4</order></author><author><firstname>Romain</firstname><surname>Janvier</surname><order>5</order></author><author><firstname>Federico</firstname><surname>Giulioni</surname><order>6</order></author><author><firstname>Miriam</firstname><surname>Herrmann</surname><order>7</order></author><author><firstname>Andrew</firstname><surname>Hudak</surname><order>8</order></author><author><firstname>Russell</firstname><surname>Parsons</surname><order>9</order></author><author><firstname>Cristina</firstname><surname>Santin</surname><order>10</order></author></authors><documents><document><filename>71694__36463__25e6fe6aebd64163aa27cd1eec724eb8.pdf</filename><originalFilename>71694.VoR.pdf</originalFilename><uploaded>2026-04-01T10:33:49.2440223</uploaded><type>Output</type><contentLength>3220825</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9;The Author(s) 2025. This is an Open Access article distributed under the terms of the Creative Commons Attribution License.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/)</licence></document></documents><OutputDurs/></rfc1807>
spelling 2026-04-01T10:36:31.2682553 v2 71694 2026-04-01 SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests 2026-04-01 Semantic segmentation of point clouds using deep learning (DL) has been the subject of research in forestry in recent years due to its potential applications. Several scientific and management disciplines, such as biodiversity monitoring, ecosystem carbon assessments, or forest management could benefit from this technique. However, it requires manual segmentation of point clouds to be used as training data. This process is highly labour-intensive and time-consuming, and there is a notable lack of publicly available datasets to support the development of accurate DL semantic segmentation models for forestry and forest ecology applications. Here, we present SegmentedForests, a curated dataset of manually segmented ground-based point clouds from forest plots, specifically designed to facilitate the training and validation of semantic segmentation models. This publicly available dataset contains >920 million labelled points from 14 forest plots, acquired using both terrestrial laser scanning (TLS) and mobile laser scanning (MLS) technologies. It covers two hectares of broadleaf, conifer, and mixed stands from different bioclimatic regions and features >1600 trees across 16 tree species. Each point cloud is labelled into multiple vegetation classes (up to 16), such as tree stems, branches, grass, shrubs, and down wood, as well as non-vegetation elements commonly present in forest scenes, including rocks, people, and stakes. Data splits to facilitate DL model development using our dataset are provided as well. The dataset is available at https://zenodo.org/records/17396681. By releasing this annotated dataset, we seek to address the critical need for publicly available, high-quality training data for DL models that perform semantic segmentation of ground-based point clouds in forest ecosystems. Journal Article Forestry: An International Journal of Forest Research 99 2 Oxford University Press (OUP) 0015-752X 1464-3626 forestry; deep learning; artificial intelligence; TLS; supervised segmentation; 3D computer vision 1 4 2026 2026-04-01 10.1093/forestry/cpaf062 A correction has been published: Forestry: An International Journal of Forest Research, Volume 99, Issue 2, April 2026, cpag018, https://doi.org/10.1093/forestry/cpag018 COLLEGE NANME COLLEGE CODE Swansea University This work was supported by the UK NERC project(NE/T001194/1): ‘Advancing 3D Fuel Mapping for Wildfire Behaviour and Risk Mitigation Modelling’, the Spanish Knowledge Generation project (PID2021-126790NB-I00): ‘Advancing carbon emission estimations from wildfires applying artificial intelligence to 3D terrestrial point clouds’, (grant PRE2022-104159) funded by MCIN/AEI/10.13039/501100011033 and FSE+, research grant ‘FIREPROs’ (IDE/2024/000780) funded by the Principality of Asturias Government (Spain), COST Action 3DForEcoTech CA20118 supported by COST (European Cooperation in Science and Technology) and the US DoD SERDP/ESTCP projects (RC201025), (RC23-7626), and (RC20-1046). 2026-04-01T10:36:31.2682553 2026-04-01T10:26:11.3754926 Faculty of Science and Engineering School of Biosciences, Geography and Physics - Geography Diego Laino 1 Carlos Cabo 2 Celestino Ordóñez 3 Rodolfo Bolanos 4 Romain Janvier 5 Federico Giulioni 6 Miriam Herrmann 7 Andrew Hudak 8 Russell Parsons 9 Cristina Santin 10 71694__36463__25e6fe6aebd64163aa27cd1eec724eb8.pdf 71694.VoR.pdf 2026-04-01T10:33:49.2440223 Output 3220825 application/pdf Version of Record true ©The Author(s) 2025. This is an Open Access article distributed under the terms of the Creative Commons Attribution License. true eng https://creativecommons.org/licenses/by/4.0/)
title SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
spellingShingle SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
,
title_short SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
title_full SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
title_fullStr SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
title_full_unstemmed SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
title_sort SegmentedForests: a labelled dataset of terrestrial LiDAR point clouds for semantic segmentation of forests
author ,
author2 Diego Laino
Carlos Cabo
Celestino Ordóñez
Rodolfo Bolanos
Romain Janvier
Federico Giulioni
Miriam Herrmann
Andrew Hudak
Russell Parsons
Cristina Santin
format Journal article
container_title Forestry: An International Journal of Forest Research
container_volume 99
container_issue 2
publishDate 2026
institution Swansea University
issn 0015-752X
1464-3626
doi_str_mv 10.1093/forestry/cpaf062
publisher Oxford University Press (OUP)
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Biosciences, Geography and Physics - Geography{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Biosciences, Geography and Physics - Geography
document_store_str 1
active_str 0
description Semantic segmentation of point clouds using deep learning (DL) has been the subject of research in forestry in recent years due to its potential applications. Several scientific and management disciplines, such as biodiversity monitoring, ecosystem carbon assessments, or forest management could benefit from this technique. However, it requires manual segmentation of point clouds to be used as training data. This process is highly labour-intensive and time-consuming, and there is a notable lack of publicly available datasets to support the development of accurate DL semantic segmentation models for forestry and forest ecology applications. Here, we present SegmentedForests, a curated dataset of manually segmented ground-based point clouds from forest plots, specifically designed to facilitate the training and validation of semantic segmentation models. This publicly available dataset contains >920 million labelled points from 14 forest plots, acquired using both terrestrial laser scanning (TLS) and mobile laser scanning (MLS) technologies. It covers two hectares of broadleaf, conifer, and mixed stands from different bioclimatic regions and features >1600 trees across 16 tree species. Each point cloud is labelled into multiple vegetation classes (up to 16), such as tree stems, branches, grass, shrubs, and down wood, as well as non-vegetation elements commonly present in forest scenes, including rocks, people, and stakes. Data splits to facilitate DL model development using our dataset are provided as well. The dataset is available at https://zenodo.org/records/17396681. By releasing this annotated dataset, we seek to address the critical need for publicly available, high-quality training data for DL models that perform semantic segmentation of ground-based point clouds in forest ecosystems.
published_date 2026-04-01T07:01:27Z
_version_ 1861428013005012992
score 11.1007185

Similar Items