EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Descriptron‐ GBIF Annotator: A browser‐based platform for crowdsourced morphological annotation of biodiversity images

Alex R. Van Dam, Francisco Hita Garcia

Methods in Ecology and Evolution · 2026-11 · 17 (11) · 3076-3083

DOI: 10.1111/2041-210x.70408 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Methods in Ecology and Evolution, online erschienen am 08.09.2026. Verfasst von Alex R. Van Dam und 1 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The biodiversity crisis demands new approaches to describing and characterising species that can scale beyond the capacity of professional taxonomists alone. Hundreds of millions of specimen images are now openly available, yet almost none carry structured information about the morphology they depict. We present the Descriptron‐GBIF Annotator, a free, zero‐installation tool that runs entirely in a web browser and lets anyone create structured, ontology‐linked morphological annotations of specimen images drawn directly from the Global Biodiversity Information Facility (GBIF) and the taxonomic literature. AI‐assisted outlining is provided by the Segment Anything Model 2 (SAM2), and a library of anatomical templates covering 25 major taxonomic groups guides the annotator to standardised body regions and traits. Annotations can be exported in widely used formats: Darwin Core, COCO, a flat traits table and a linked‐data (JSON‐LD) knowledge graph, and published directly to Zenodo as citable, FAIR datasets with DOIs. We position the tool as the public‐facing first tier of a two‐tier system that complements a professional workbench (the Descriptron Portal) used by taxonomists, creating a feedback loop in which public annotations help train AI models and expert‐validated outputs improve the public tool. In roughly 5 months of quiet availability the tool has logged 2248 usage events, including 280 AI‐assisted segmentations across 43 anatomical regions with high mask confidence (median 0.92). We describe the tool, walk through a complete worked example, report this early usage and discuss what wider adoption will require. The annotator is available at https://www.descriptrongbifannotator.org and the source code at https://doi.org/10.5281/zenodo.18888577 and https://github.com/alexrvandam/Descriptron‐GBIF_Annotator .

Zitieren

APA

Dam, A. R. V., & Garcia, F. H. (2026). Descriptron‐ GBIF Annotator: A browser‐based platform for crowdsourced morphological annotation of biodiversity images. Methods in Ecology and Evolution, 17(11), 3076-3083. https://doi.org/10.1111/2041-210x.70408

Vancouver

Dam ARV, Garcia FH. Descriptron‐ GBIF Annotator: A browser‐based platform for crowdsourced morphological annotation of biodiversity images. Methods in Ecology and Evolution. 2026;17(11):3076-3083. doi:10.1111/2041-210x.70408

BibTeX
@article{Dam2026,
  author = {Dam, Alex R. Van and Garcia, Francisco Hita},
  title = {Descriptron‐ GBIF Annotator: A browser‐based platform for crowdsourced morphological annotation of biodiversity images},
  journal = {Methods in Ecology and Evolution},
  year = {2026},
  volume = {17},
  number = {11},
  pages = {3076--3083},
  doi = {10.1111/2041-210x.70408}
}
Kostenlos starten →