Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Dec 18;25(24):7678.
doi: 10.3390/s25247678.

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

Partho Ghose  1 , Al Bashir  1 , Yibin Wang  1 , Cristian Bua  1   2 , Azlan Zahid  1
Affiliations

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

Partho Ghose et al. Sensors (Basel). .

Abstract

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.

Keywords: SAM; YOLO; detection; few-shot; precision agriculture; segmentation; strawberries; zero-shot.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
Sample images from datasets: (a) an image of a greenhouse-grown strawberry plant with some hanging fruits and (b) an image of field-grown strawberry plants with some fruits lying on the soil bed.
Figure 2
Figure 2
Detailed architecture of YOLOv11 model.
Figure 3
Figure 3
Schematic diagram of YOLO-SAM AgriScan framework for ripe strawberry segmentation.
Figure 4
Figure 4
Effect of training epochs on detection performance across datasets D1 and D2.
Figure 5
Figure 5
PR curves for ripe strawberry detection for different epoch numbers (a) for D1 and (b) D2 datasets.
Figure 6
Figure 6
Bar chart comparing mean Dice and IoU scores for ripe strawberry segmentation on Dataset1(D1) and Dataset2(D2).
Figure 7
Figure 7
Qualitative assessment on two datasets applying YOLOv11-MobileSAM and YOLOv11-SAM with YOLO-SAM AgriScan.
Figure 8
Figure 8
Missed target detection and segmentation case in YOLO-SAM AgriScan.
See this image and copyright information in PMC

References

    1. Yang Q., Liu L., Zhou J., Rogers M., Jin Z. Predicting the growth trajectory and yield of greenhouse strawberries based on knowledge-guided computer vision. Comput. Electron. Agric. 2024;220:108911. doi: 10.1016/j.compag.2024.108911. - DOI
    1. Júnior M.R.B., de Almeida Moreira B.R., dos Santos Carreira V., de Brito Filho A.L., Trentin C., de Souza F.L.P., Tedesco D., Setiyono T., Flores J.P., Ampatzidis Y., et al. Precision agriculture in the United States: A comprehensive meta-review inspiring further research, innovation, and adoption. Comput. Electron. Agric. 2024;221:108993. doi: 10.1016/j.compag.2024.108993. - DOI
    1. Bai Y., Yu J., Yang S., Ning J. An improved YOLO algorithm for detecting flowers and fruits on strawberry seedlings. Biosyst. Eng. 2024;237:1–12. doi: 10.1016/j.biosystemseng.2023.11.008. - DOI
    1. Bashir A., Ojo M., Zahid A. Real-time Estimation of Strawberry Maturity Level and Count Using CNN in Controlled Environment Agriculture; Proceedings of the 2023 ASABE Annual International Meeting; Omaha, Nebraska. 8–12 July 2023; St. Joseph, MI, USA: American Society of Agricultural and Biological Engineers; 2023. p. 1.
    1. He Z., Khanal S.R., Zhang X., Karkee M., Zhang Q. Real-time strawberry detection based on improved yolov5s architecture for robotic harvesting in open-field environment. arXiv. 20232308.03998

LinkOut - more resources