WINMOL Analyzer
Detection and quantification of windthrown tree stems on UAV-orthomosaics

Severe storm events are the biggest driving factor for biomass loss in European forests. Besides the damages caused by the storm itself, there are subsequent damages due to biotic, abiotic and market factors with far reaching implications for forestry and conservation. These subsequent damages can be minimized if the amount and spatial distribution of the windthrown trees is known and can be used to optimize salvage operations and calamity management.

Traditional methods and space born remote sensing can only provide estimations of the affected area, whereas remote sensing with aerial sensors is able to obtain the spatial distribution of the stems with high detection rates up but not to quantify single logs due to an insufficient spatial resolution. 

The WINMOL Analyzer closes this gap by analyzing UAV-orthomosaics with Deep Learning techniques to obtain precise information of the spatial distribution and estimate the volume of the windthrown trees. Therefore, the U-Net was included in a bottom-up object detection based on a skeletonization algorithm including a reconstruction of occluded stem parts by a voting system based on morphological heuristics. For the subsequent quantification of the detected stems, the diameter is determined every 25 cm, and the volume is calculated as sum of truncated cone volumes.

For the development of WINMOL Analyzer, 21 orthomosaics of beech, spruce and mixed stands with a ground sampling distance of mostly less than 2 cm were used, on which 1747 windthrown tree stems were manually outlined for the training of the provide deep learning models “Spruce”, “Beech” and “General” for pure and mixed stands, respectively. Additionally, 710 trees were digitized and measured for the validation of the methodology.

It could be proven that the proposed methodology is able to detect windthrown tree stems with an average detection rate of 93.8% for spruce stands and with 93.3% for beech and mixed stans (error rates between 4.3% and 6.3%). Thereby, the detected volumes were overestimated by the specific models for spruce (11.8%) and beech stands (10.7%) while the “General” model was underestimating the volume by 5.8%, on average. Generally, specialization on certain tree species carries the risk of lower detection rates for unfamiliar scenes and species, while a general model is associated with a higher rate of classification errors. Further, it could be shown that the performance of the proposed methodology is affected by the quality of the used orthomosaics.

The quantification of the amount of windthrown wood and the reconstruction of occluded stem parts are unique features compared to other recently published approaches with far reaching implications for forestry, ecology, and biodiversity conservation. The proposed methodology provides additional information for decision-making in the planning of salvage loggings and for monitoring of biomass and carbon cycles. Further, it can contribute to a better understanding of windthrow dynamics and therewith, will support the development of sustainable management strategies which focus on resilient forest ecosystems.

For detailed information please visit the original paper which is available as preprint.

Windthrow damages caused by a the winter storm in January 2022 in North-East Germany


UAV Orthomosaic


The WINMOL Analyzer is optimized for the analysis of UAV-orthomosaics with a ground sampling distance of less than 3 cm.
Semantic Stem Map

Semantic Stem Map

In the first step, a deep learning model is utilized to calculate a semantic stem map which highlights the stem pixels.

Detected Stems

Stem Detection

Stems segments are detected based on a skeletonization algorithm and merged  based on a heuristic voting system. Thereby, also occluded stem parts are reconstructed.
Stem Quantification

Stem Quantification

The volumes of the stems are determined by measuring the stem diameter every 25 cm. The stem volume is calculated as sum of truncated cones.
WINMOL Analyzer User Interface

WINMOL Analyzer

Welcome to the WINMOL Analyzer – the comprehensive solution for detecting and quantifying windthrown tree stems on UAV-orthomosaics!
Leveraging the deep learning-based segmentation and a unique heraldic voting system for stem reconstruction, the WINMOL Analyzer is a robust and automated tool for identifying wind-damaged trees, measuring stem dimensions, and estimating wood volume.

Simplify your analysis and elevate decision-making in forest management with the WINMOL Analyzer, leading the forefront in UAV-based deep learning for forestry.

#WindthrowAalysis #ForestryTech #DeepLearning #Forest4.0 

How to use the WINMOL Analyzer

The WINMOL Analyzer is an easy to use QGis Plugin with a user-friendly GUI. You are only a few steps away from your first analyzed windthrow damage!
Input UAV Orthomosaic

Choose an UAV-Orthomosaic of the windthrow area which should be analyzed. It is recommended to use orthomosaics with a pixel size of less than 3 cm to get optimal results as the provided deep learning models were trained on this pixel size. If the pixel size is significantly greater, the detection rate will decrease dramatically.


Choose a model for the semantic segmentation of the UAV-orthomosaics. Tree species specific models are provided for beech and spruce and a general model for mixed stands or other tree species. Also, custom models can be used.
For the training of custom models please use the WINMOL Segmentor.

[Optional] Options for Semantic Segmentation

These options can only be changed if a custom model is chosen. It is necessary to set it according to the extend of the used training tiles to ensure a proper classification if the used model architecture is not scale invariant.
Tile side length describes the extend of the used training tiles in m.
Tile pixel extend describes the extend of the used training tiles in pixel.

Options for Stem Detection

With this parameters, the set of heuristics used in the stem detection and reconstruction process can be adapted.
Min Length defines the lower threshold of the length of a stem to be recognized.
Max Distance limits the distance between two corresponding stem segments to be merged in the stem reconstruction process. This threshold should be set according to the maximum tree crown diameter of the analyzed stand.
Maximum Angle between two corresponding stem segments to be reconstructed.
Maximum Tree Height defines an upper threshold for the maximum stem length in the reconstruction process. Higher threshold will slow down the stem reconstruction process. Set it close to the maximum tree height to get ideal results.


Choose the output according to the needs!
Detected windthrown trees generates a line vector layer of the detected tree stems which can be used
Measuring nodes contains point vectors of all measuring nodes and their respective diameters and measuring angle.
Semantic Stem Map exports the output of the deep learning model as raster layer.

Installation Guide

Download the Plugin.
Open QGis.
Go to Plugins -> Manage and install Plugins.
  Use Install from ZIP to install the downloaded archive.
The installation may takes several minutes!


Visit on
WIMOL Podcast - "Die Überflieger"
Sturmschäden im Wald verstehen 
Januar 6, 2024

Stefan Reder and Nicole Albert are talking about the reasons for storm damages in forests.
 - only available in German

Visit on
WIMOL Podcast - "Die Überflieger"
Sturmschäden aus der Vogelperspektive 
Januar 6, 2024

Stefan Reder and Nicole Albert are presenting the developed methodology for the detection of storm damages in forest. 
 - only available in German

Visti on
WIMOL Podcast - "Die Überflieger"
“Modellierung von Sturmschäden“ 
Januar 6, 2024

Line Grottian and Catrin Stadelmann are presenting their research results on modelling of storm risks in forests.
 - only available in German

Mobirise Website Builder
Article Published:
Detection of Windthrown Tree Stems on UAV-Orthomosaics Using U-Net Convolutional Networks
Stefan Reder, Jan-Peter Mund, Nicole Albert, Lilli Waßermann und Luis Miranda 2022

The increasing number of severe storm events is threatening European forests. Besides the primary damages directly caused by storms, there are secondary damages such as bark beetle outbreaks and tertiary damages due to negative effects on the market. These subsequent damages can be minimized if a detailed overview of the affected area and the amount of damaged wood can be obtained quickly and included in the planning of clearance measures. The present work utilizes UAV-orthophotos and an adaptation of the U-Net architecture for the semantic segmentation and localization of windthrown stems. The network was pre-trained with generic datasets, randomly combining stems and background samples in a copy–paste augmentation, and afterwards trained with a specific dataset of a particular windthrow. The models pre-trained with generic datasets containing 10, 50 and 100 augmentations per annotated windthrown stems achieved F1-scores of 73.9% (S1Mod10), 74.3% (S1Mod50) and 75.6% (S1Mod100), outperforming the baseline model (F1-score 72.6%), which was not pre-trained. These results emphasize the applicability of the method to correctly identify windthrown trees and suggest the collection of training samples from other tree species and windthrow areas to improve the ability to generalize. Further enhancements of the network architecture are considered to improve the classification performance and to minimize the calculative costs.

Mobirise Website Builder
WINMOL Project Homepage
July 2020 - December 2023

WINMIOL is a joined project of the Eberswalde University for Sustainable development and the Thünen Institute for Exosystem Research Eberswalde for the analysis and modeling of storm damages in forests. 



Eberswalde University for Sustainable Development
Departement of Forest and Environment
Alfred-Möller-Str. 1
16225 Eberswalde, Germany

The WINMOL Analyzer was created in the scope of the WINMOL project,  a shared project of the Eberswalde University for Sustainable Development and the Thünen Institute of Forest Ecosystem Eberswalde, funded by the Fachagentur für Nachwachsende Rohstoffe e.V. (FNR) and the Federal Ministry of Food and Agriculture, Germany (BMEL).
FNR funding number: 2220NR024A
Copyright © 2023 Hochschule für nachhaltige Entwicklung Eberswalde | mundialis GmbH & Co. KG| Terrestris GmbH & Co. KG

Mobirise Website Builder
Mobirise Website Builder
Mobirise Website Builder
Mobirise Website Builder
Mobirise Website Builder
Mobirise Website Builder

AI Website Software