Deep Learning Black-Box and Pattern Recognition Analysis Using Guided Grad-CAM for Phytolith Identification.

Journal: Annals of botany
Published Date: May 30, 2025

Abstract

BACKGROUND AND AIMS: In this article, visual explainers are applied to give transparency to the black-box of a trained VGG19 model for the identification of multi-cell phytoliths of the Avena, Hordeum and Triticum genera. The aim is to demonstrate its proper learning by visually highlighting the phytolith characteristics that the deep learning model uses to classify these phytoliths, we then compare the model's methods to those employed manually by archaeobotanists.

Authors

  • Iban Berganzo-Besga
    Computational Social Sciences and Humanities Department, Barcelona Supercomputing Center (BSC-CNS), Plaça d'Eusebi Güell, 1-3, Les Corts, 08034, Barcelona, Spain.
  • Hector A Orengo
    Computational Social Sciences and Humanities Department, Barcelona Supercomputing Center (BSC-CNS), Plaça d'Eusebi Güell, 1-3, Les Corts, 08034, Barcelona, Spain.
  • Felipe Lumbreras
    Computer Vision Center (CVC), 08193, Bellaterra (Cerdanyola del Vallès), Spain.
  • Monica N Ramsey
    Ramsey Laboratory for Environmental Archaeology (RLEA), Department of Anthropology, University of Toronto Mississauga (UTM), 3359 Mississauga Road, Mississauga, ON, L5L 1C6, Canada.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals