EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automated extraction of leaf mass per area from digitized herbarium specimens

Thais Vasconcelos iD, William N Weaver iD, Aly Baumgartner iD, Zoë Bugnaski iD, James Boyko iD

New Phytologist · 2026-07 · 251 (2) · 896-908

DOI: 10.1111/nph.70292 ↗

Kurz gefasst

Fachartikel in der Zeitschrift New Phytologist, online erschienen am 18.06.2025. Verfasst von Thais Vasconcelos und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI, 48109, USA. Hauptthemen laut MEDLINE: Plant Leaves, Magnoliopsida, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: BioXFEL Science and Technology Center.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The digitization of vast herbarium collections has made millions of plant specimen images freely available online, which can now be used to generate phenotypic datasets of unprecedented scope. Here, we assess the potential of computer vision tools to automate the extraction of predicted leaf mass per area (LMA pred ) from digitized herbarium specimens. We use an automated pipeline to extract leaf area and petiole width from 22 680 leaves, representing a phylogenetic informed sample of 1580 species of woody angiosperms. LMA pred is estimated using a proxy equation that models the scaling relationship between petiole width and leaf mass. We assess potential sources of error in LMA pred estimates and evaluate whether documented LMA–climate patterns are recovered using this dataset and phylogenetic comparative methods. Our LMA pred dataset responds mainly to temperature and solar radiation and presents a positive correlation with latitude. The proxy equation, not the automated pipeline, is responsible for most of the error in LMA pred estimates. Our pipeline underscores the power of combining herbarium digitization with new techniques for automated trait scoring. The increased size of datasets generated using this tool allows investigation of potential LMA–climate relationships with a geographically balanced sample while also utilizing comprehensive phylogenetic information.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Plant LeavesMagnoliopsidaImage Processing, Computer-AssistedAutomationPhylogenyClimateOrgan SizeTemperature

Stichwörter der Autor:innen

angiospermsleaf arealeaf economic spectrumleaf massmachine learningprecipitationsolar radiation

Autor:innen und Institutionen

  • Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI, 48109, USA.
  • University of Michigan Herbarium, University of Michigan, Ann Arbor, MI, 48108, USA.
  • Michigan Institute of Data Science, University of Michigan, Ann Arbor, MI, 48109, USA.

ORCID-Profile: Thais Vasconcelos, William N Weaver, Aly Baumgartner, Zoë Bugnaski, James Boyko

Förderung

  • BioXFEL Science and Technology Center

Zitieren

APA

Vasconcelos, T., Weaver, W. N., Baumgartner, A., Bugnaski, Z., & Boyko, J. (2026). Automated extraction of leaf mass per area from digitized herbarium specimens. New Phytologist, 251(2), 896-908. https://doi.org/10.1111/nph.70292

Vancouver

Vasconcelos T, Weaver WN, Baumgartner A, Bugnaski Z, Boyko J. Automated extraction of leaf mass per area from digitized herbarium specimens. New Phytologist. 2026;251(2):896-908. doi:10.1111/nph.70292

BibTeX
@article{Vasconcelos2026,
  author = {Vasconcelos, Thais and Weaver, William N and Baumgartner, Aly and Bugnaski, Zoë and Boyko, James},
  title = {Automated extraction of leaf mass per area from digitized herbarium specimens},
  journal = {New Phytologist},
  year = {2026},
  volume = {251},
  number = {2},
  pages = {896--908},
  doi = {10.1111/nph.70292}
}
Kostenlos starten →