EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

AT‐ViT: Area‐Targeted Multi‐View Vision Transformer With Cross‐Attention and Multi‐Scale Patching for Plant Trait Recognition in Herbarium Images

Amani Sedrat, Takieddine Chehhat, Youcef Sklab, Hanane Ariouat, Abderrazak Sebaa, Eric Chenin, Jean‐Daniel Zucker, Edi Prifti

IET Computer Vision · 2026-01 · 20 (1)

DOI: 10.1049/cvi2.70059 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IET Computer Vision, online erschienen am 14.03.2026. Verfasst von Amani Sedrat und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Agence Nationale de la Recherche.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Automated plant traits recognition from herbarium images is essential for plant sciences, yet it remains challenging because background elements (e.g., textual labels, mounting artefacts and colour charts) can introduce shortcut learning, leading models to rely on spurious nonplant cues rather than plant morphology. This bias degrades both generalisation and interpretability. In this paper, we introduce AT‐ViT , a dual‐branch vision transformer that jointly encodes raw herbarium scans and their segmented‐derived counterparts via a multi‐scale, multi‐view cross‐attention fusion scheme. AT‐ViT further incorporates a mask‐guided patch weighting mechanism that amplifies plant‐relevant regions and attenuates background‐driven features. By learning from the original scans while being guided by segmentation masks through the mask‐guided patch reweighting mechanism, the model is encouraged to focus on plant organs and learn plant‐centric representations more effectively. Across multiple trait classification tasks (e.g., leaf base shape, thorns), AT‐ViT delivers consistent accuracy gains, improves attention localisation on plant regions and exhibits increased robustness under synthetic background perturbations. Specifically, AT‐ViT substantially improves spatial attention grounding, boosting plant‐region alignment (Avg IoU_p: +15.66 to +18.03 pp) while reducing background overlap (Avg IoU_b: −27.92 to −31.02 pp) relative to CrossViT, and remains markedly more robust to background perturbations, outperforming ResNet101 by up to +32.32 accuracy points and CrossViT by up to +5.07 points under background‐noise conditions.

Förderung

  • Agence Nationale de la Recherche

Zitieren

APA

Sedrat, A., Chehhat, T., Sklab, Y., Ariouat, H., Sebaa, A., Chenin, E., Zucker, J., & Prifti, E. (2026). AT‐ViT: Area‐Targeted Multi‐View Vision Transformer With Cross‐Attention and Multi‐Scale Patching for Plant Trait Recognition in Herbarium Images. IET Computer Vision, 20(1). https://doi.org/10.1049/cvi2.70059

Vancouver

Sedrat A, Chehhat T, Sklab Y, Ariouat H, Sebaa A, Chenin E, et al. AT‐ViT: Area‐Targeted Multi‐View Vision Transformer With Cross‐Attention and Multi‐Scale Patching for Plant Trait Recognition in Herbarium Images. IET Computer Vision. 2026;20(1). doi:10.1049/cvi2.70059

BibTeX
@article{Sedrat2026,
  author = {Sedrat, Amani and Chehhat, Takieddine and Sklab, Youcef and Ariouat, Hanane and Sebaa, Abderrazak and Chenin, Eric and Zucker, Jean‐Daniel and Prifti, Edi},
  title = {AT‐ViT: Area‐Targeted Multi‐View Vision Transformer With Cross‐Attention and Multi‐Scale Patching for Plant Trait Recognition in Herbarium Images},
  journal = {IET Computer Vision},
  year = {2026},
  volume = {20},
  number = {1},
  doi = {10.1049/cvi2.70059}
}
Kostenlos starten →