• GOPI M
• kumaraswamy s

Pomegranate cultivation in India faces challenges in growth stage monitoring, impacting yield and resource efficiency. This paper proposes a YOLOv11-based model to automate pomegranate growth stage detection (bud, flower, early fruit, mid- growth, ripe) using high-resolution images. The model achieves a mean average precision (mAP@50) of 0.875 and mAP@50- 95 of 0.723, outperforming manual methods. Trained on 5,858 annotated images (80% training, 10% validation/testing), the system integrates data augmentation and hyperparameter tun- ing for robustness. Results demonstrate its potential for pre- cision agriculture, enabling optimized irrigation, pest control, and harvesting. Experimental results show that the model achieves mAP@0.5 of 87.5% and mAP@0.5–0.95 of 72.3%, with high real-time inference capability. Visual analysis through confidence-threshold and precision-recall curves further validates the model’s robustness. This system offers a scalable, fast, and reliable solution for real-time monitoring and decision-making in pomegranate farming. Its integration with IoT-based platforms can significantly aid farmers in optimizing irrigation, fertilization, pest control, and harvesting schedules—ultimately improving yield quality and reducing losses.

Pomegranate growth stages, YOLOv11, preci- sion agriculture, deep learning, object detection

"Deep Learning-Based Pomegranate Growth Stage Detection Using YOLOv11 for Precision Agriculture", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.k823-k830, May-2025, Available :http://www.jetir.org/papers/JETIR2505B73.pdf

"Deep Learning-Based Pomegranate Growth Stage Detection Using YOLOv11 for Precision Agriculture", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppk823-k830, May-2025, Available at : http://www.jetir.org/papers/JETIR2505B73.pdf