EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

YOLO-SAM AgriScan: A Unified Framework for Ripe Strawberry Detection and Segmentation with Few-Shot and Zero-Shot Learning

Partho Ghose iD, Al Bashir iD, Yibin Wang, Cristian Bua iD, Azlan Zahid iD

Sensors · 18.12.2025 · 25 (24) · 7678

DOI: 10.3390/s25247678 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 18.12.2025. Verfasst von Partho Ghose und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biological and Agricultural Engineering, Texas A&M AgriLife Research, Texas A&M University System, Dallas, TX 75252, USA. Hauptthemen laut MEDLINE: Fragaria, Image Processing, Computer-Assisted, Fruit. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional segmentation methods are slow and rely on manual annotations, which are labor-intensive. To address these limitations, we propose YOLO-SAM AgriScan, a unified framework that combines the fast object detection capabilities of YOLOv11 with the zero-shot segmentation power of the Segment Anything Model 2 (SAM2). Our approach adopts a hybrid paradigm for on-plant ripe strawberry segmentation, wherein YOLOv11 is fine-tuned using a few-shot learning strategy with minimal annotated samples, and SAM2 performs mask generation without additional supervision. This architecture eliminates the bottleneck of pixel-wise manual annotation and enables the scalable and efficient segmentation of strawberries in both controlled and natural farm environments. Experimental evaluations on two datasets, a custom-collected dataset and a publicly available benchmark, demonstrate strong detection and segmentation performance in both full-data and data-constrained scenarios. The proposed framework achieved a mean Dice score of 0.95 and an IoU of 0.93 on our collected dataset and maintained competitive performance on public data (Dice: 0.95, IoU: 0.92), demonstrating its robustness, generalizability, and practical relevance in real-world agricultural settings. Our results highlight the potential of combining few-shot detection and zero-shot segmentation to accelerate the development of annotation-light, intelligent phenotyping systems.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

FragariaImage Processing, Computer-AssistedFruitAlgorithmsMachine LearningAgriculture

Stichwörter der Autor:innen

SAMYOLOdetectionfew-shotprecision agriculturesegmentationstrawberrieszero-shot

Autor:innen und Institutionen

  • Department of Biological and Agricultural Engineering, Texas A&M AgriLife Research, Texas A&M University System, Dallas, TX 75252, USA.
  • Department of Information Engineering, University of Pisa, 56122 Pisa, Italy.

ORCID-Profile: Partho Ghose, Al Bashir, Cristian Bua, Azlan Zahid

Zitieren

APA

Ghose, P., Bashir, A., Wang, Y., Bua, C., & Zahid, A. (2025). YOLO-SAM AgriScan: A Unified Framework for Ripe Strawberry Detection and Segmentation with Few-Shot and Zero-Shot Learning. Sensors, 25(24), 7678. https://doi.org/10.3390/s25247678

Vancouver

Ghose P, Bashir A, Wang Y, Bua C, Zahid A. YOLO-SAM AgriScan: A Unified Framework for Ripe Strawberry Detection and Segmentation with Few-Shot and Zero-Shot Learning. Sensors. 2025;25(24):7678. doi:10.3390/s25247678

BibTeX
@article{Ghose2025,
  author = {Ghose, Partho and Bashir, Al and Wang, Yibin and Bua, Cristian and Zahid, Azlan},
  title = {YOLO-SAM AgriScan: A Unified Framework for Ripe Strawberry Detection and Segmentation with Few-Shot and Zero-Shot Learning},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {24},
  pages = {7678},
  doi = {10.3390/s25247678}
}
Kostenlos starten →