EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Improving micromorphological analysis with CNN-based segmentation of flint/obsidian, bone and charcoal

Rafael Arnay, Pedro García-Villa iD, Javier Hernández-Aceituno iD, Sara Rueda-Saiz iD, Carolina Mallol

PLOS One · 2026

DOI: 10.1371/journal.pone.0340353 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Rafael Arnay und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Engineering and Systems, University of La Laguna, La Laguna, Santa Cruz de Tenerife, Spain. Hauptthemen laut MEDLINE: Charcoal, Bone and Bones, Archaeology. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Agencia Estatal de Investigación.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The quantification and identification of components in archaeological micromorphology remain subjective and challenging, particularly for early-career researchers. To address this, we developed a deep learning tool for the automatic segmentation of three materials commonly found in Palaeolithic contexts and thin sections: bone, charcoal, and lithic fine-grained debitage (flint and obsidian). Using high-resolution photomicrographs of 57 thin sections in plane-polarised and cross-polarised light, we trained and evaluated state-of-the-art convolutional neural networks (CNNs) for material segmentation. The best-performing configuration, a U-Net with an InceptionV4 encoder, achieved mean intersection over union (IoU) scores of 0.96 for flint/obsidian, 0.80 for bone, and 0.82 for charcoal. The models also classified the relative abundance of each material with balanced accuracies of 0.99 for flint/obsidian, 0.92 for bone, and 0.85 for charcoal. These results demonstrate the potential of deep learning to enhance objectivity, accuracy, and reproducibility in archaeological micromorphology, providing a valuable resource for future geoarchaeological research.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CharcoalBone and BonesArchaeologyConvolutional Neural NetworksFossilsAnimalsDeep LearningNeural Networks, Computer

Substanzen

Charcoal

Autor:innen und Institutionen

  • Department of Computer Engineering and Systems, University of La Laguna, La Laguna, Santa Cruz de Tenerife, Spain.
  • Archaeological Micromorphology and Biomarker Research Lab, Instituto Universitario de Bio-Orgánica Antonio González (IUBO), La Laguna, Santa Cruz de Tenerife, Spain.

ORCID-Profile: Pedro García-Villa, Javier Hernández-Aceituno, Sara Rueda-Saiz

Förderung

  • Agencia Estatal de Investigación

Zitieren

APA

Arnay, R., García-Villa, P., Hernández-Aceituno, J., Rueda-Saiz, S., & Mallol, C. (2026). Improving micromorphological analysis with CNN-based segmentation of flint/obsidian, bone and charcoal. PLOS One. https://doi.org/10.1371/journal.pone.0340353

Vancouver

Arnay R, García-Villa P, Hernández-Aceituno J, Rueda-Saiz S, Mallol C. Improving micromorphological analysis with CNN-based segmentation of flint/obsidian, bone and charcoal. PLOS One. 2026. doi:10.1371/journal.pone.0340353

BibTeX
@article{Arnay2026,
  author = {Arnay, Rafael and García-Villa, Pedro and Hernández-Aceituno, Javier and Rueda-Saiz, Sara and Mallol, Carolina},
  title = {Improving micromorphological analysis with CNN-based segmentation of flint/obsidian, bone and charcoal},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0340353}
}
Kostenlos starten →