EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

From blender to farm: Transforming controlled environment agriculture with synthetic data and SwinUNet for precision crop monitoring

Kimia Aghamohammadesmaeilketabforoosh iD, Joshua Parfitt, Soodeh Nikan, Joshua M Pearce iD

PLOS One · 24.04.2025 · 20 (4) · e0322189

DOI: 10.1371/journal.pone.0322189 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 24.04.2025. Verfasst von Kimia Aghamohammadesmaeilketabforoosh und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical & Computer Engineering, Western University, London, Ontario, Canada. Hauptthemen laut MEDLINE: Fragaria, Crops, Agricultural, Agriculture. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Weston Family Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The aim of this study was to train a Vision Transformer (ViT) model for semantic segmentation to differentiate between ripe and unripe strawberries using synthetic data to avoid challenges with conventional data collection methods. The solution used Blender to generate synthetic strawberry images along with their corresponding masks for precise segmentation. Subsequently, the synthetic images were used to train and evaluate the SwinUNet as a segmentation method, and Deep Domain Confusion was utilized for domain adaptation. The trained model was then tested on real images from the Strawberry Digital Images dataset. The performance on the real data achieved a Dice Similarity Coefficient of 94.8% for ripe strawberries and 94% for unripe strawberries, highlighting its effectiveness for applications such as fruit ripeness detection. Additionally, the results show that increasing the volume and diversity of the training data can significantly enhance the segmentation accuracy of each class. This approach demonstrates how synthetic datasets can be employed as a cost-effective and efficient solution for overcoming data scarcity in agricultural applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

FragariaCrops, AgriculturalAgricultureFruitFarmsAlgorithms

Autor:innen und Institutionen

  • Department of Electrical & Computer Engineering, Western University, London, Ontario, Canada.
  • College of Engineering, Pennsylvania State University, University Park, Pennsylvania, United States of America.
  • Ivey Business School, Western University, London, Ontario, Canada.

ORCID-Profile: Kimia Aghamohammadesmaeilketabforoosh, Joshua M Pearce

Förderung

  • Weston Family Foundation

Zitieren

APA

Aghamohammadesmaeilketabforoosh, K., Parfitt, J., Nikan, S., & Pearce, J. M. (2025). From blender to farm: Transforming controlled environment agriculture with synthetic data and SwinUNet for precision crop monitoring. PLOS One, 20(4), e0322189. https://doi.org/10.1371/journal.pone.0322189

Vancouver

Aghamohammadesmaeilketabforoosh K, Parfitt J, Nikan S, Pearce JM. From blender to farm: Transforming controlled environment agriculture with synthetic data and SwinUNet for precision crop monitoring. PLOS One. 2025;20(4):e0322189. doi:10.1371/journal.pone.0322189

BibTeX
@article{Aghamohammadesmaeilketabforoosh2025,
  author = {Aghamohammadesmaeilketabforoosh, Kimia and Parfitt, Joshua and Nikan, Soodeh and Pearce, Joshua M},
  title = {From blender to farm: Transforming controlled environment agriculture with synthetic data and SwinUNet for precision crop monitoring},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {4},
  pages = {e0322189},
  doi = {10.1371/journal.pone.0322189}
}
Kostenlos starten →