EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Deep learning to capture leaf shape in plant images: Validation by geometric morphometrics

Ladislav Hodač iD, Kevin Karbstein, Lara Kösters, Michael Rzanny, Hans Christian Wittich, David Boho, David Šubrt, Patrick Mäder, Jana Wäldchen

The Plant Journal · 2024-11 · 120 (4) · 1343-1357

DOI: 10.1111/tpj.17053 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Plant Journal, online erschienen am 09.10.2024. Verfasst von Ladislav Hodač und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department Biogeochemical Integration, Max Planck Institute for Biogeochemistry, Jena, Germany. Hauptthemen laut MEDLINE: Plant Leaves, Deep Learning, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 109 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: German Federal Ministry of Education and Research (BMBF), German Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection (BMUV), Thuringian Ministry for Environment, Energy and Nature Conservation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Plant leaves play a pivotal role in automated species identification using deep learning (DL). However, achieving reproducible capture of leaf variation remains challenging due to the inherent “black box” problem of DL models. To evaluate the effectiveness of DL in capturing leaf shape, we used geometric morphometrics (GM), an emerging component of eXplainable Artificial Intelligence (XAI) toolkits. We photographed Ranunculus auricomus leaves directly in situ and after herbarization. From these corresponding leaf images, we automatically extracted DL features using a neural network and digitized leaf shapes using GM. The association between the extracted DL features and GM shapes was then evaluated using dimension reduction and covariation models. DL features facilitated the clustering of leaf images by source populations in both in situ and herbarized leaf image datasets, and certain DL features were significantly associated with biological leaf shape variation as inferred by GM. DL features also enabled leaf classification into morpho‐phylogenomic groups within the intricate R. auricomus species complex. We demonstrated that simple in situ leaf imaging and DL reproducibly captured leaf shape variation at the population level, while combining this approach with GM provided key insights into the shape information extracted from images by computer vision, a necessary prerequisite for reliable automated plant phenotyping.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Plant LeavesDeep LearningImage Processing, Computer-AssistedNeural Networks, ComputerPhylogeny

Stichwörter der Autor:innen

Ranunculus auricomusdeep learningeXplainable AIgeometric morphometricsleaf imagesphenotypic variationsmartphone imaging in situ

Autor:innen und Institutionen

  • Department Biogeochemical Integration, Max Planck Institute for Biogeochemistry, Jena, Germany.
  • Data-intensive Systems and Visualization Group, Technische Universität Ilmenau, Ilmenau, Germany.
  • Faculty of Science, Jan Evangelista Purkyně University in Ústí nad Labem, Ústí nad Labem, Czech Republic.
  • German Centre for Integrative Biodiversity Research - iDiv (Halle-Jena-Leipzig), Leipzig, Germany.
  • Faculty of Biological Sciences, Friedrich Schiller University Jena, Jena, Germany.

ORCID-Profile: Ladislav Hodač

Förderung

  • German Federal Ministry of Education and Research (BMBF)
  • German Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection (BMUV)
  • Thuringian Ministry for Environment, Energy and Nature Conservation

Zitieren

APA

Hodač, L., Karbstein, K., Kösters, L., Rzanny, M., Wittich, H. C., Boho, D., Šubrt, D., Mäder, P., & Wäldchen, J. (2024). Deep learning to capture leaf shape in plant images: Validation by geometric morphometrics. The Plant Journal, 120(4), 1343-1357. https://doi.org/10.1111/tpj.17053

Vancouver

Hodač L, Karbstein K, Kösters L, Rzanny M, Wittich HC, Boho D, et al. Deep learning to capture leaf shape in plant images: Validation by geometric morphometrics. The Plant Journal. 2024;120(4):1343-1357. doi:10.1111/tpj.17053

BibTeX
@article{Hoda2024,
  author = {Hodač, Ladislav and Karbstein, Kevin and Kösters, Lara and Rzanny, Michael and Wittich, Hans Christian and Boho, David and Šubrt, David and Mäder, Patrick and Wäldchen, Jana},
  title = {Deep learning to capture leaf shape in plant images: Validation by geometric morphometrics},
  journal = {The Plant Journal},
  year = {2024},
  volume = {120},
  number = {4},
  pages = {1343--1357},
  doi = {10.1111/tpj.17053}
}
Kostenlos starten →