Benchmarking tree species classification from proximally sensed laser scanning data: introducing the FOR-species20K dataset

Simple item page
dc.contributor.authorPulti, Stefano
dc.contributor.authorLines, Emily R.
dc.contributor.authorMullerova, Jana
dc.contributor.authorFrey, Julian
dc.contributor.authorSchindler, Zoe
dc.contributor.authorStraker, Adrian
dc.contributor.authorAllen, Matthew J.
dc.contributor.authorWiniwarter, Lukas
dc.contributor.authorRehush, Nataliia
dc.contributor.authorHristova, Hristina
dc.contributor.authorMurray, Brent
dc.contributor.authorCalders, Kim
dc.contributor.authorCoops, Nicholas
dc.contributor.authorHofle, Bernhard
dc.contributor.authorIrwin, Liam
dc.contributor.authorJunttila, Samuli
dc.contributor.authorKrucek, Martin
dc.contributor.authorKrok, Grzegorz
dc.contributor.authorKral, Kamil
dc.contributor.authorLevick, Shaun R.
dc.contributor.authorLuck, Linda
dc.contributor.authorMissarov, Azim
dc.contributor.authorMokros, Martin
dc.contributor.authorOwen, Harry J. F.
dc.contributor.authorSterenczak, Krzysztof
dc.contributor.authorPitkanen, Timo P.
dc.contributor.authorPuletti, Nicola
dc.contributor.authorSaarinen, Ninni
dc.contributor.authorHopkinson, Christopher
dc.contributor.authorTerryn, Louise
dc.contributor.authorTorresan, Chiara
dc.contributor.authorTomelleri, Enrico
dc.contributor.authorWeiser, Hannah
dc.contributor.authorAstrup, Rasmus
dc.date.accessioned2025-06-24T23:11:27Z
dc.date.available2025-06-24T23:11:27Z
dc.date.issued2025
dc.descriptionOpen access article. Creative Commons Attribution-NonCommercial 4.0 International license (CC BY-NC 4.0) applies
dc.description.abstract1. Proximally sensed laser scanning presents new opportunities for automated forest ecosystem data capture. However, a gap remains in deriving ecologically pertinent information, such as tree species, without additional ground data. Artificial intelligence approaches, particularly deep learning (DL), have shown promise towards automation. Progress has been limited by the lack of large, diverse, and, most importantly, openly available labelled single-tree point cloud datasets. This has hindered both (1) the robustness of the DL models across varying data types (platforms and sensors) and (2) the ability to effectively track progress, thereby slowing the convergence towards best practice for species classification. 2. To address the above limitations, we compiled the FOR-species20K benchmark dataset, consisting of individual tree point clouds captured using proximally sensed laser scanning data from terrestrial (TLS), mobile (MLS) and drone laser scanning (ULS). Compiled collaboratively, the dataset includes data collected in forests mainly across Europe, covering Mediterranean, temperate and boreal biogeographic regions. It includes scattered tree data from other continents, totaling over 20,000 trees of 33 species and covering a wide range of tree sizes and forms. Alongside the release of FOR-species20K, we benchmarked seven leading DL models for individual tree species classification, including both point cloud (PointNet++, MinkNet, MLP-Mixer, DGCNNs) and multi-view 2D-based methods (SimpleView, DetailView, YOLOv5). 3. 2D Image-based models had, on average, higher overall accuracy (0.77) than 3D point cloud-based models (0.72). Notably, the performance was consistently >0.8 across scanning platforms and sensors, offering versatility in deployment. The top-scoring model, DetailView, demonstrated robustness to training data imbalances and effectively generalized across tree sizes. 4. The FOR-species20K dataset represents an important asset for developing and benchmarking DL models for individual tree species classification using proximally sensed laser scanning data. As such, it serves as a crucial foundation for future efforts to classify accurately and map tree species at various scales using laser scanning technology, as it provides the complete code base, dataset, and an initial baseline representative of the current state-of-the-art of point cloud tree species classification methods.
dc.description.peer-reviewYes
dc.identifier.citationPulti, S., Lines, E. R., Mullerova, J., Frey, J., Schindler, Z., Straker, A., Allen, M. J., Winiwarter, L., Rehush, N., Hristova, H., Murray, B., Calders., K., Coops, N., Hofle, B., Irwin. L., Junttila, S., Krucek, M., Krok, G., Kral, K.,...Astrup. R. (2025). Benchmarking tree species classification from proximally sensed laser scanning data: Introducing the FOR-species20K dataset. Methods in Ecology and Evolution, 16(4), 801-818. https://doi.org/10.1111/2041-210X.14503
dc.identifier.urihttps://hdl.handle.net/10133/7061
dc.language.isoen
dc.publisherWiley
dc.publisher.departmentDepartment of Geography and Environment
dc.publisher.facultyArts and Science
dc.publisher.institutionNorwegian Institute for Bioeconomy Research (NIBIO)
dc.publisher.institutionUniversity of Cambridge
dc.publisher.institutionJan Evangelista Purkyne University
dc.publisher.institutionUniversity of Freiburg
dc.publisher.institutionGeorg-August-Universitat Gottingen
dc.publisher.institutionTU Wien
dc.publisher.institutionSwiss Federal Institute for Forest, Snow and Landscape Research WSL
dc.publisher.institutionUniversity of British Columbia
dc.publisher.institutionGhent University
dc.publisher.institutionHeidelberg University
dc.publisher.institutionUniversity of Eastern Finland
dc.publisher.institutionSilva Tarouca Research Institute
dc.publisher.institutionForest Research Institute (Poland)
dc.publisher.institutionCommonwealth Scientific and Industrial Research Organization (CSIRO)
dc.publisher.institutionCharles Darwin University
dc.publisher.institutionGerman Research Centre for GeoSciences (GFZ)
dc.publisher.institutionUniversity College of London
dc.publisher.institutionNatural Resources Institute Finland (Luke)
dc.publisher.institutionResearch Centre for Forestry and Wood (Italy)
dc.publisher.institutionUniversity of Lethbridge
dc.publisher.institutionInstitute of BioEconomy (Italy)
dc.publisher.institutionFree University of Bolzano
dc.publisher.urlhttps://doi.org/10.1111/2041-210X.14503
dc.subjectBiodiversity
dc.subjectDeep learning
dc.subjectLidar
dc.subjectPoint cloud classification
dc.subjectRemote sensing
dc.subjectSingle-tree inventory
dc.subjectForest ecosystem
dc.subjectTree data
dc.subjectTree species
dc.subject.lcshTrees--Classification
dc.subject.lcshForest ecology
dc.titleBenchmarking tree species classification from proximally sensed laser scanning data: introducing the FOR-species20K dataset
dc.typeArticle
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Hopkinson-benchmarking-tree.pdf
Size:
10.64 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.75 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Hopkinson, Christopher