Pubblicazioni Scientifiche

Filtri di ricerca 2 risultati
Reset
Pubblicazioni per anno
Benchmarking tree species classification from proximally sensed laser scanning data: Introducing the FOR-species20K dataset
Puliti , Stefano , Lines , Emily R. , Müllerová , Jana , Frey , Julian , Schindler , Zoe , Straker , Adrian , Allen , Matthew J. , Winiwarter , Lukas , Rehush , Nataliia , Hristova , Hristina S. , Murray , Brent A. , Calders , Kim , Coops , Nicholas C. , Höfle , Bernhard , Irwin , Liam A.K. , Junttila , Samuli , Kruček , Martin , Krok , G. , Král , Kamil , Levick , Shaun R. , Lück , Linda , Missarov , Azim , Mokroš , M. , Owen , Harry Jon Foord , Stereńczak , Krzysztof Jan , Pitkänen , Timo P. , Puletti , Nicola , Saarinen , Ninni , Hopkinson , Chris Dennis , Terryn , Louise , Torresan , C. , Tomelleri , Enrico , Weiser , Hannah , Astrup , Rasmus
lidar remote sensing biodiversity deep learning point cloud classification single-tree inventory
Mostra abstract
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. 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). 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. 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. © 2025 The Author(s). Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society.
🔗 Vai alla pubblicazione
2025
Rivista
Methods in Ecology and Evolution
Dettagli
Vol. 16 (4) pp. 801 - 818
DOI
Handbook of field sampling for multi-taxon biodiversity studies in European forests
Burrascano , Sabina , Trentanovi , Giovanni , Paillet , Yoan , Heilmann-Clausen , Jacob , Giordani , P. , Bagella , Simonetta , Bravo-Oviedo , Andrés , Campagnaro , Thomas , Campanaro , Alessandro , Chianucci , Francesco , de Smedt , Pallieter , Itziar , García Mijangos , Matošević , Dinka , Sitzia , Tommaso , Aszalós , Réka , Brazaitis , Gediminas , Cutini , Andrea , D'Andrea , Ettore , Doerfler , Inken , Hofmeister , Jeňýk , Hošek , Jan , Janssen , Philippe , Kepfer-Rojas , Sebastian , Korboulewsky , Nathalie , Kozák , Daniel , Lachat , Thibault , Lõhmus , Asko , López , Rosana , Mårell , Anders , Matula , Radim , Mikoláš , Martin , Munzi , Silvana , Nordén , Björn , Pärtel , Meelis , Penner , Johannes , Runnel , Kadri , Schall , Peter , Svoboda , Miroslav , Tinya , Flóra , Ujházyová , Mariana , Vandekerkhove , Kris , Verheyen , Kris , Xystrakis , Fotios , Ódor , Péter
biodiversity forest stand structure multi-taxon field methods indicators sampling protocol
Mostra abstract
Forests host most terrestrial biodiversity and their sustainable management is crucial to halt biodiversity loss. Although scientific evidence indicates that sustainable forest management (SFM) should be assessed by monitoring multi-taxon biodiversity, most current SFM criteria and indicators account only for trees or consider indirect biodiversity proxies. Several projects performed multi-taxon sampling to investigate the effects of forest management on biodiversity, but the large variability of their sampling approaches hampers the identification of general trends, and limits broad-scale inference for designing SFM. Here we address the need of common sampling protocols for forest structure and multi-taxon biodiversity to be used at broad spatial scales. We established a network of researchers involved in 41 projects on forest multi-taxon biodiversity across 13 European countries. The network data structure comprised the assessment of at least three taxa, and the measurement of forest stand structure in the same plots or stands. We mapped the sampling approaches to multi-taxon biodiversity, standing trees and deadwood, and used this overview to provide operational answers to two simple, yet crucial, questions: what to sample? How to sample? The most commonly sampled taxonomic groups are vascular plants (83% of datasets), beetles (80%), lichens (66%), birds (66%), fungi (61%), bryophytes (49%). They cover different forest structures and habitats, with a limited focus on soil, litter and forest canopy. Notwithstanding the common goal of assessing forest management effects on biodiversity, sampling approaches differed widely within and among taxonomic groups. Differences derive from sampling units (plots size, use of stand vs. plot scale), and from the focus on different substrates or functional groups of organisms. Sampling methods for standing trees and lying deadwood were relatively homogeneous and focused on volume calculations, but with a great variability in sampling units and diameter thresholds. We developed a handbook of sampling methods (SI 3) aimed at the greatest possible comparability across taxonomic groups and studies as a basis for European-wide biodiversity monitoring programs, robust understanding of biodiversity response to forest structure and management, and the identification of direct indicators of SFM. © 2021 The Authors
🔗 Vai alla pubblicazione
2021
Rivista
Ecological Indicators
Dettagli
Vol. 132
DOI