A deep-learning framework for enhancing habitat identification based on species composition

César Leblanc; Pierre Bonnet; Maximilien Servajean; Milan Chytrý; Svetlana Aćić; Olivier Argagnon; Ariel Bergamini; Idoia Biurrun; Gianmaria Bonari; Juan A. Campos; Andraž Čarni; Renata Ćušterevska; Michele De Sanctis; Jürgen Dengler; Emmanuel Garbolino; Valentin Golub; Ute Jandt; Florian Jansen; Maria Lebedeva; Jonathan Lenoir; Jesper Erenskjold Moeslund; Aaron Pérez-Haase; Remigiusz Pielech; Jozef Šibík; Zvjezdana Stančić; Angela Stanisci; Grzegorz Swacha; Domas Uogintas; Kiril Vassilev; Thomas Wohlgemuth; Alexis Joly

doi:10.1111/avsc.12802

A deep-learning framework for enhancing habitat identification based on species composition

César Leblanc^*, Pierre Bonnet, Maximilien Servajean, Milan Chytrý, Svetlana Aćić, Olivier Argagnon, Ariel Bergamini, Idoia Biurrun, Gianmaria Bonari, Juan A. Campos, Andraž Čarni, Renata Ćušterevska, Michele De Sanctis, Jürgen Dengler, Emmanuel Garbolino, Valentin Golub, Ute Jandt, Florian Jansen, Maria Lebedeva, Jonathan LenoirJesper Erenskjold Moeslund, Aaron Pérez-Haase, Remigiusz Pielech, Jozef Šibík, Zvjezdana Stančić, Angela Stanisci, Grzegorz Swacha, Domas Uogintas, Kiril Vassilev, Thomas Wohlgemuth, Alexis Joly

^*Corresponding author for this work

Department of Ecoscience

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Abstract

Aims: The accurate classification of habitats is essential for effective biodiversity conservation. The goal of this study was to harness the potential of deep learning to advance habitat identification in Europe. We aimed to develop and evaluate models capable of assigning vegetation-plot records to the habitats of the European Nature Information System (EUNIS), a widely used reference framework for European habitat types. Location: The framework was designed for use in Europe and adjacent areas (e.g., Anatolia, Caucasus). Methods: We leveraged deep-learning techniques, such as transformers (i.e., models with attention components able to learn contextual relations between categorical and numerical features) that we trained using spatial k-fold cross-validation (CV) on vegetation plots sourced from the European Vegetation Archive (EVA), to show that they have great potential for classifying vegetation-plot records. We tested different network architectures, feature encodings, hyperparameter tuning and noise addition strategies to identify the optimal model. We used an independent test set from the National Plant Monitoring Scheme (NPMS) to evaluate its performance and compare its results against the traditional expert systems. Results: Exploration of the use of deep learning applied to species composition and plot-location criteria for habitat classification led to the development of a framework containing a wide range of models. Our selected algorithm, applied to European habitat types, significantly improved habitat classification accuracy, achieving a more than twofold improvement compared to the previous state-of-the-art (SOTA) method on an external data set, clearly outperforming expert systems. The framework is shared and maintained through a GitHub repository. Conclusions: Our results demonstrate the potential benefits of the adoption of deep learning for improving the accuracy of vegetation classification. They highlight the importance of incorporating advanced technologies into habitat monitoring. These algorithms have shown to be better suited for habitat type prediction than expert systems. They push the accuracy score on a database containing hundreds of thousands of standardized presence/absence European surveys to 88.74%, as assessed by expert judgment. Finally, our results showcase that species dominance is a strong marker of ecosystems and that the exact cover abundance of the flora is not required to train neural networks with predictive performances. The framework we developed can be used by researchers and practitioners to accurately classify habitats.

Original language	English
Article number	e12802
Journal	Applied Vegetation Science
Volume	27
Issue	3
ISSN	1402-2001
DOIs	https://doi.org/10.1111/avsc.12802
Publication status	Published - Jul 2024

Keywords

artificial intelligence
biodiversity monitoring
deep learning
European flora
expert system
habitat type identification
phytosociology
species composition
vascular plants
vegetation classification

Access to Document

10.1111/avsc.12802Licence: CC BY

Cite this

Leblanc, C., Bonnet, P., Servajean, M., Chytrý, M., Aćić, S., Argagnon, O., Bergamini, A., Biurrun, I., Bonari, G., Campos, J. A., Čarni, A., Ćušterevska, R., De Sanctis, M., Dengler, J., Garbolino, E., Golub, V., Jandt, U., Jansen, F., Lebedeva, M., ... Joly, A. (2024). A deep-learning framework for enhancing habitat identification based on species composition. Applied Vegetation Science, 27(3), Article e12802. https://doi.org/10.1111/avsc.12802

@article{6f04ae44efc04b569c7888c7e190de49,

title = "A deep-learning framework for enhancing habitat identification based on species composition",

abstract = "Aims: The accurate classification of habitats is essential for effective biodiversity conservation. The goal of this study was to harness the potential of deep learning to advance habitat identification in Europe. We aimed to develop and evaluate models capable of assigning vegetation-plot records to the habitats of the European Nature Information System (EUNIS), a widely used reference framework for European habitat types. Location: The framework was designed for use in Europe and adjacent areas (e.g., Anatolia, Caucasus). Methods: We leveraged deep-learning techniques, such as transformers (i.e., models with attention components able to learn contextual relations between categorical and numerical features) that we trained using spatial k-fold cross-validation (CV) on vegetation plots sourced from the European Vegetation Archive (EVA), to show that they have great potential for classifying vegetation-plot records. We tested different network architectures, feature encodings, hyperparameter tuning and noise addition strategies to identify the optimal model. We used an independent test set from the National Plant Monitoring Scheme (NPMS) to evaluate its performance and compare its results against the traditional expert systems. Results: Exploration of the use of deep learning applied to species composition and plot-location criteria for habitat classification led to the development of a framework containing a wide range of models. Our selected algorithm, applied to European habitat types, significantly improved habitat classification accuracy, achieving a more than twofold improvement compared to the previous state-of-the-art (SOTA) method on an external data set, clearly outperforming expert systems. The framework is shared and maintained through a GitHub repository. Conclusions: Our results demonstrate the potential benefits of the adoption of deep learning for improving the accuracy of vegetation classification. They highlight the importance of incorporating advanced technologies into habitat monitoring. These algorithms have shown to be better suited for habitat type prediction than expert systems. They push the accuracy score on a database containing hundreds of thousands of standardized presence/absence European surveys to 88.74%, as assessed by expert judgment. Finally, our results showcase that species dominance is a strong marker of ecosystems and that the exact cover abundance of the flora is not required to train neural networks with predictive performances. The framework we developed can be used by researchers and practitioners to accurately classify habitats.",

keywords = "artificial intelligence, biodiversity monitoring, deep learning, European flora, expert system, habitat type identification, phytosociology, species composition, vascular plants, vegetation classification",

author = "C{\'e}sar Leblanc and Pierre Bonnet and Maximilien Servajean and Milan Chytr{\'y} and Svetlana A{\'c}i{\'c} and Olivier Argagnon and Ariel Bergamini and Idoia Biurrun and Gianmaria Bonari and Campos, {Juan A.} and Andra{\v z} {\v C}arni and Renata {\'C}u{\v s}terevska and Michele De Sanctis and J{\"u}rgen Dengler and Emmanuel Garbolino and Valentin Golub and Ute Jandt and Florian Jansen and Maria Lebedeva and Jonathan Lenoir and Moeslund, {Jesper Erenskjold} and Aaron P{\'e}rez-Haase and Remigiusz Pielech and Jozef {\v S}ib{\'i}k and Zvjezdana Stan{\v c}i{\'c} and Angela Stanisci and Grzegorz Swacha and Domas Uogintas and Kiril Vassilev and Thomas Wohlgemuth and Alexis Joly",

note = "Publisher Copyright: {\textcopyright} 2024 The Author(s). Applied Vegetation Science published by John Wiley & Sons Ltd on behalf of International Association for Vegetation Science.",

year = "2024",

month = jul,

doi = "10.1111/avsc.12802",

language = "English",

volume = "27",

journal = "Applied Vegetation Science",

issn = "1402-2001",

publisher = "Wiley-Blackwell",

number = "3",

}

Leblanc, C, Bonnet, P, Servajean, M, Chytrý, M, Aćić, S, Argagnon, O, Bergamini, A, Biurrun, I, Bonari, G, Campos, JA, Čarni, A, Ćušterevska, R, De Sanctis, M, Dengler, J, Garbolino, E, Golub, V, Jandt, U, Jansen, F, Lebedeva, M, Lenoir, J, Moeslund, JE, Pérez-Haase, A, Pielech, R, Šibík, J, Stančić, Z, Stanisci, A, Swacha, G, Uogintas, D, Vassilev, K, Wohlgemuth, T & Joly, A 2024, 'A deep-learning framework for enhancing habitat identification based on species composition', Applied Vegetation Science, vol. 27, no. 3, e12802. https://doi.org/10.1111/avsc.12802

A deep-learning framework for enhancing habitat identification based on species composition. / Leblanc, César; Bonnet, Pierre; Servajean, Maximilien et al.
In: Applied Vegetation Science, Vol. 27, No. 3, e12802, 07.2024.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

TY - JOUR

T1 - A deep-learning framework for enhancing habitat identification based on species composition

AU - Leblanc, César

AU - Bonnet, Pierre

AU - Servajean, Maximilien

AU - Chytrý, Milan

AU - Aćić, Svetlana

AU - Argagnon, Olivier

AU - Bergamini, Ariel

AU - Biurrun, Idoia

AU - Bonari, Gianmaria

AU - Campos, Juan A.

AU - Čarni, Andraž

AU - Ćušterevska, Renata

AU - De Sanctis, Michele

AU - Dengler, Jürgen

AU - Garbolino, Emmanuel

AU - Golub, Valentin

AU - Jandt, Ute

AU - Jansen, Florian

AU - Lebedeva, Maria

AU - Lenoir, Jonathan

AU - Moeslund, Jesper Erenskjold

AU - Pérez-Haase, Aaron

AU - Pielech, Remigiusz

AU - Šibík, Jozef

AU - Stančić, Zvjezdana

AU - Stanisci, Angela

AU - Swacha, Grzegorz

AU - Uogintas, Domas

AU - Vassilev, Kiril

AU - Wohlgemuth, Thomas

AU - Joly, Alexis

PY - 2024/7

Y1 - 2024/7

N2 - Aims: The accurate classification of habitats is essential for effective biodiversity conservation. The goal of this study was to harness the potential of deep learning to advance habitat identification in Europe. We aimed to develop and evaluate models capable of assigning vegetation-plot records to the habitats of the European Nature Information System (EUNIS), a widely used reference framework for European habitat types. Location: The framework was designed for use in Europe and adjacent areas (e.g., Anatolia, Caucasus). Methods: We leveraged deep-learning techniques, such as transformers (i.e., models with attention components able to learn contextual relations between categorical and numerical features) that we trained using spatial k-fold cross-validation (CV) on vegetation plots sourced from the European Vegetation Archive (EVA), to show that they have great potential for classifying vegetation-plot records. We tested different network architectures, feature encodings, hyperparameter tuning and noise addition strategies to identify the optimal model. We used an independent test set from the National Plant Monitoring Scheme (NPMS) to evaluate its performance and compare its results against the traditional expert systems. Results: Exploration of the use of deep learning applied to species composition and plot-location criteria for habitat classification led to the development of a framework containing a wide range of models. Our selected algorithm, applied to European habitat types, significantly improved habitat classification accuracy, achieving a more than twofold improvement compared to the previous state-of-the-art (SOTA) method on an external data set, clearly outperforming expert systems. The framework is shared and maintained through a GitHub repository. Conclusions: Our results demonstrate the potential benefits of the adoption of deep learning for improving the accuracy of vegetation classification. They highlight the importance of incorporating advanced technologies into habitat monitoring. These algorithms have shown to be better suited for habitat type prediction than expert systems. They push the accuracy score on a database containing hundreds of thousands of standardized presence/absence European surveys to 88.74%, as assessed by expert judgment. Finally, our results showcase that species dominance is a strong marker of ecosystems and that the exact cover abundance of the flora is not required to train neural networks with predictive performances. The framework we developed can be used by researchers and practitioners to accurately classify habitats.

AB - Aims: The accurate classification of habitats is essential for effective biodiversity conservation. The goal of this study was to harness the potential of deep learning to advance habitat identification in Europe. We aimed to develop and evaluate models capable of assigning vegetation-plot records to the habitats of the European Nature Information System (EUNIS), a widely used reference framework for European habitat types. Location: The framework was designed for use in Europe and adjacent areas (e.g., Anatolia, Caucasus). Methods: We leveraged deep-learning techniques, such as transformers (i.e., models with attention components able to learn contextual relations between categorical and numerical features) that we trained using spatial k-fold cross-validation (CV) on vegetation plots sourced from the European Vegetation Archive (EVA), to show that they have great potential for classifying vegetation-plot records. We tested different network architectures, feature encodings, hyperparameter tuning and noise addition strategies to identify the optimal model. We used an independent test set from the National Plant Monitoring Scheme (NPMS) to evaluate its performance and compare its results against the traditional expert systems. Results: Exploration of the use of deep learning applied to species composition and plot-location criteria for habitat classification led to the development of a framework containing a wide range of models. Our selected algorithm, applied to European habitat types, significantly improved habitat classification accuracy, achieving a more than twofold improvement compared to the previous state-of-the-art (SOTA) method on an external data set, clearly outperforming expert systems. The framework is shared and maintained through a GitHub repository. Conclusions: Our results demonstrate the potential benefits of the adoption of deep learning for improving the accuracy of vegetation classification. They highlight the importance of incorporating advanced technologies into habitat monitoring. These algorithms have shown to be better suited for habitat type prediction than expert systems. They push the accuracy score on a database containing hundreds of thousands of standardized presence/absence European surveys to 88.74%, as assessed by expert judgment. Finally, our results showcase that species dominance is a strong marker of ecosystems and that the exact cover abundance of the flora is not required to train neural networks with predictive performances. The framework we developed can be used by researchers and practitioners to accurately classify habitats.

KW - artificial intelligence

KW - biodiversity monitoring

KW - deep learning

KW - European flora

KW - expert system

KW - habitat type identification

KW - phytosociology

KW - species composition

KW - vascular plants

KW - vegetation classification

UR - http://www.scopus.com/inward/record.url?scp=85202490569&partnerID=8YFLogxK

U2 - 10.1111/avsc.12802

DO - 10.1111/avsc.12802

M3 - Journal article

AN - SCOPUS:85202490569

SN - 1402-2001

VL - 27

JO - Applied Vegetation Science

JF - Applied Vegetation Science

IS - 3

M1 - e12802

ER -

A deep-learning framework for enhancing habitat identification based on species composition

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this