EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Dynamic LoRA Fine-Tuning of DINOv3 for Multi-Component Pasture Biomass Estimation

Shikha Sen, Nischay Dhankhar, Akram Bayat

Sensors · 20.08.2026 · 26 (16) · 5285

DOI: 10.3390/s26165285 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 20.08.2026. Verfasst von Shikha Sen und 2 weiteren Autor:innen. Beteiligte Einrichtung: Khoury College of Computer Sciences, Northeastern University, Boston, MA 02115, USA. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate estimation of pasture biomass components from imagery is essential for sustainable grazing management and precision agriculture. Conventional methods such as destructive harvesting, rising plate meters, and remote sensing are limited by scalability, reliability, or the ability to disaggregate biomass by species. We propose a parameter-efficient multi-output regression framework predicting five biomass components (dry green, dry dead, dry clover, green dry matter, and total dry biomass) from high-resolution top-view pasture images. It employs a pretrained DINOv3 Vision Transformer backbone adapted via a dynamic, depth-aware Low-Rank Adaptation (LoRA) strategy, in which the adaptation rank and scaling factor increase exponentially with layer depth: early layers encoding generic visual primitives are minimally perturbed, while deeper layers receive stronger task-specific adaptation. This schedule is effective in low-data regimes, where uniform adaptation or full fine-tuning overfits. To handle rectangular image geometry, each image is split into two square halves processed as a dual-view stream with a contrastive alignment loss. The system ensembles ViT-Large and ViT-Huge backbones with test-time augmentation across five-fold cross-validation. On the CSIRO Image2Biomass benchmark, the full pipeline attains a cross-validated weighted R-squared of 0.81, indicating that depth-aware, parameter-efficient adaptation of large vision models is effective for non-invasive biomass estimation under data scarcity.

Schlagwörter

Stichwörter der Autor:innen

DINOv3Low-Rank AdaptationVision Transformermulti-output regressionparameter-efficient fine-tuningpasture biomass estimationprecision agriculture

Autor:innen und Institutionen

  • Khoury College of Computer Sciences, Northeastern University, Boston, MA 02115, USA.
  • Nace AI, Palo Alto, CA 94306, USA.

Zitieren

APA

Sen, S., Dhankhar, N., & Bayat, A. (2026). Dynamic LoRA Fine-Tuning of DINOv3 for Multi-Component Pasture Biomass Estimation. Sensors, 26(16), 5285. https://doi.org/10.3390/s26165285

Vancouver

Sen S, Dhankhar N, Bayat A. Dynamic LoRA Fine-Tuning of DINOv3 for Multi-Component Pasture Biomass Estimation. Sensors. 2026;26(16):5285. doi:10.3390/s26165285

BibTeX
@article{Sen2026,
  author = {Sen, Shikha and Dhankhar, Nischay and Bayat, Akram},
  title = {Dynamic LoRA Fine-Tuning of DINOv3 for Multi-Component Pasture Biomass Estimation},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {16},
  pages = {5285},
  doi = {10.3390/s26165285}
}
Kostenlos starten →