EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Self-supervised adaptation of vision foundation models for fine-grained plant species and damage assessment in herbicide trials

Leire Benito-Del-Valle, Artzai Picón, Daniel Mugica, Manuel Ramos, Eva Portillo, Javier Romero, Carlos Javier Jimenez, Ramón Navarra-Mestre

Computers and Electronics in Agriculture · 2027-02 · 257 · 112572

DOI: 10.1016/j.compag.2026.112572 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers and Electronics in Agriculture, 2027-02. Verfasst von Leire Benito-Del-Valle und 7 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Basque Government.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Herbicide field trials require accurate identification of plant species and assessment of herbicide-induced damage across diverse environments. While general-purpose vision foundation models have shown promising results in complex visual domains, their performance can be limited in agriculture, where fine-grained distinctions between species and damage types are critical. In this work, we adapt DINOv2, a general-purpose vision foundation model, to herbicide trial characterization. Using self-supervised learning on 6 million image tiles from 340,554 herbicide-trial images spanning multiple years, locations, sensors, and crop species, the model learns representations optimized for herbicide trial images. Our domain-specific model significantly outperforms DINOv2 in both species identification (F1 score from 0.91 to 0.94) and damage classification (from 0.26 to 0.33) under in-domain conditions. Under unseen conditions (new locations and times), it achieves greater gains (species identification from 0.56 to 0.66; damage classification from 0.17 to 0.27), although absolute damage performance remains relatively low. In domain-shift scenarios, such as drone imagery, species classification improves from 0.49 to 0.60. We show that domain-specific pretraining enhances segmentation accuracy, particularly in low-annotation regimes. An annotation-efficiency analysis reveals that, under unseen conditions, the domain-specific model achieves 5.4% higher F1 score than DINOv2, while using 80% fewer labeled samples. On the public PhenoBench benchmark, which addresses a simpler crop/weed/soil segmentation task, a modest global F1 gain (0.736 to 0.742) indicates more limited evidence for cross-task transfer. These results demonstrate that large-scale, domain-specific self-supervised pretraining significantly improves in-domain performance and robustness under domain shift, while reducing manual annotation efforts, and providing a foundation for scalable and automated herbicide trial analysis.

Förderung

  • Basque Government

Zitieren

APA

Benito-Del-Valle, L., Picón, A., Mugica, D., Ramos, M., Portillo, E., Romero, J., Jimenez, C. J., & Navarra-Mestre, R. (2027). Self-supervised adaptation of vision foundation models for fine-grained plant species and damage assessment in herbicide trials. Computers and Electronics in Agriculture, 257, 112572. https://doi.org/10.1016/j.compag.2026.112572

Vancouver

Benito-Del-Valle L, Picón A, Mugica D, Ramos M, Portillo E, Romero J, et al. Self-supervised adaptation of vision foundation models for fine-grained plant species and damage assessment in herbicide trials. Computers and Electronics in Agriculture. 2027;257:112572. doi:10.1016/j.compag.2026.112572

BibTeX
@article{BenitoDelValle2027,
  author = {Benito-Del-Valle, Leire and Picón, Artzai and Mugica, Daniel and Ramos, Manuel and Portillo, Eva and Romero, Javier and Jimenez, Carlos Javier and Navarra-Mestre, Ramón},
  title = {Self-supervised adaptation of vision foundation models for fine-grained plant species and damage assessment in herbicide trials},
  journal = {Computers and Electronics in Agriculture},
  year = {2027},
  volume = {257},
  pages = {112572},
  doi = {10.1016/j.compag.2026.112572}
}
Kostenlos starten →