Loading...
Pubblicazioni Scientifiche
Filtri di ricerca 2 risultati
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
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.
An intensity, image-based method to estimate gap fraction, canopy openness and effective leaf area index from phase-shift terrestrial laser scanning
Grotti
,
Mirko
,
Calders
,
Kim
,
Origo
,
Niall
,
Puletti
,
Nicola
,
Alivernini
,
Alessandro
,
Ferrara
,
Carlotta
,
Chianucci
,
Francesco
Mostra abstract
Accurate in situ estimates of leaf area index (LAI) are essential for a wide range of ecological studies and applications. Due to the destructiveness and impracticality of direct measurements, indirect optical methods have mostly been used in the field to derive estimates of LAI from gap fraction measurements. Terrestrial laser scanning (TLS) is strongly supporting use of this active technology, which possesses several advantages compared to passive sensors. However, edge effects and partial beam interceptions are significantly challenges for the accurate retrieval of gap fraction from 3D point cloud data available from TLS, particularly in phase-shift instruments, which in turns require point cloud filtering to correct erroneous point measurements. As the limitations above influences the point cloud, we proposed a new method which is based only on the laser return intensity (LRI) information derived from raw TLS data, which are used to generate 2D intensity images. The intensity image contains all the unfiltered LRI information captured by TLS, which is used to separate gap from non-gap pixels, using a procedure comparable to the standard image analysis processing of digital hemispherical images. This allows a theoretically consistent comparison between active and passive optical measurements of gap fraction across all the zenith angle range. The method was tested in real and simulated forests. Gap fraction, canopy openness and effective leaf area index derived from real and simulated intensity TLS images were compared with those obtained using digital hemispherical photography (DHP). Results indicated that the intensity, image-based method outperformed DHP, as the higher pixel resolution of the intensity images and the larger distance covered by TLS allowed detection of many small canopy elements, particularly at higher zenith angles (longer optical distance), which are not detected in DHP. The main findings support the reliability of the intensity, image-based method to standardize protocols for TLS phase-shift scan data processing and use of the produced canopy estimates as a benchmark for passive optical measurements. © 2019 Elsevier B.V.