Global, multi-scale standing deadwood segmentation in centimeter-scale aerial images

With tree mortality rates rising across many regions of the world, efficient methods to map dead trees are becoming increasingly important to monitor forest dieback, assess ecological impacts, and guide management strategies. Deep learning-based pattern recognition combined with the high spatial detail of aerial images from drones or airplanes provides an avenue for mapping dead tree crowns or partial canopy dieback, collectively referred to as standing deadwood. However, current methods for mapping standing deadwood are limited to specific biomes or image resolutions. Here, we present a transformer-based semantic segmentation model that generalizes across forest biomes and a wide range of image resolutions (1–28 cm) for mapping both dead tree crowns and partial canopy dieback. Our approach combines a SegFormer-based transformer architecture for image feature extraction and Focal Tversky Loss to mitigate class imbalance. We used a globally distributed crowd-sourced dataset of 434 high-resolution aerial images and manual delineations of standing deadwood of vastly varying quality. The orthophotos span all major forest biomes and cover 10,778 hectares. To further mitigate imbalances across biomes, resolutions, deadwood occurrence, and image sources, we developed a four-dimensional sampling scheme that ensures balanced representation during training. The models were trained and evaluated using heterogeneous crowd-sourced data, which, as expected, negatively affects the F1-scores. A visual inspection on independent data highlights the very precise quality of the segmentation. Our analysis revealed resolution-dependent performance variations across biomes, suggesting a relationship between optimal mapping resolution and biome-specific characteristics. We make both our model and a machine-learning-ready dataset publicly available on deadtrees.earth to support future research in tree mortality mapping.

Authors

Institutions

Publication Details

Journal
ISPRS Open Journal of Photogrammetry and Remote Sensing
Published
2025-10-22
DOI
https://doi.org/10.1016/j.ophoto.2025.100104
Primary Topic
Forest Ecology and Biodiversity Studies
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Global, multi-scale standing deadwood segmentation in centimeter-scale aerial images

Mirela Beloiu, Clemens Mosig, Myriam Cloutier, Teja Kattenborn et al.
ISPRS Open Journal of Photogrammetry and Remote Sensing
Forest Ecology and Biodiversity Studies
article

Global, multi-scale standing deadwood segmentation in centimeter-scale aerial images

Mirela Beloiu, Clemens Mosig, Myriam Cloutier, Teja Kattenborn, Martin Denter, Michael Maroschek, Julian Frey, Tommaso Jucker, Miguel D. Mahecha, Yan Pei Cheng, Samuli Junttila, Stefan Kruse, Anna Göritz, Felix Schiefer, Óscar Pérez‐Priego, Pratima Khatri‐Chhetri, Alastair Potts, Kirill Korznikov, Stéphanie Horion, Katarzyna Zielewska-Büttner, Étienne Laliberté, Rupert Seidl, Janusch Vajna-Jehle, Jakobus Möhring, Jonathan Hempel, Matthias Gassilloud, Paul Neumeier
article en

Abstract

With tree mortality rates rising across many regions of the world, efficient methods to map dead trees are becoming increasingly important to monitor forest dieback, assess ecological impacts, and guide management strategies. Deep learning-based pattern recognition combined with the high spatial detail of aerial images from drones or airplanes provides an avenue for mapping dead tree crowns or partial canopy dieback, collectively referred to as standing deadwood. However, current methods for mapping standing deadwood are limited to specific biomes or image resolutions. Here, we present a transformer-based semantic segmentation model that generalizes across forest biomes and a wide range of image resolutions (1–28 cm) for mapping both dead tree crowns and partial canopy dieback. Our approach combines a SegFormer-based transformer architecture for image feature extraction and Focal Tversky Loss to mitigate class imbalance. We used a globally distributed crowd-sourced dataset of 434 high-resolution aerial images and manual delineations of standing deadwood of vastly varying quality. The orthophotos span all major forest biomes and cover 10,778 hectares. To further mitigate imbalances across biomes, resolutions, deadwood occurrence, and image sources, we developed a four-dimensional sampling scheme that ensures balanced representation during training. The models were trained and evaluated using heterogeneous crowd-sourced data, which, as expected, negatively affects the F1-scores. A visual inspection on independent data highlights the very precise quality of the segmentation. Our analysis revealed resolution-dependent performance variations across biomes, suggesting a relationship between optimal mapping resolution and biome-specific characteristics. We make both our model and a machine-learning-ready dataset publicly available on deadtrees.earth to support future research in tree mortality mapping.

ISPRS Open Journal of Photogrammetry and Remote SensingVol. 18
University of Copenhagen (DK), Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung (DE), University of British Columbia (CA), University of Freiburg (DE), University of Eastern Finland (FI), University of Washington (US), ETH Zurich (CH), University of Bristol (GB), Institute of Botany of the Slovak Academy of Sciences (SK), Finland University (FI), Czech Academy of Sciences, Institute of Botany (CZ), Schön Klinik Berchtesgadener Land (DE), GeoInformation (United Kingdom) (GB), Technical University of Munich (DE), Université de Montréal (CA), Leipzig University (DE)
National Science Foundation, European Space Agency, Deutsche Forschungsgemeinschaft, Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Grantová Agentura České Republiky
Openalex Percentile: Top 72%
Forest Ecology and Biodiversity Studies
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.