Project information
- Category: Pomegranate Maturation Assesment
- Data: Henan dataset
- Project date: July, 2023
- Project URL: Conferenec Proceedings
YOLO-V7 BASED DETECTION AND CLASSIFICATION OF POMEGRANATE FRUITS IN DIFFERENT GROWING STAGES
In the context of Industry 4.0, this study uses the YOLO-v7 model to detect and classify pomegranates at various growth stages, enhancing harvesting efficiency and post-harvest processes. The classification includes five stages: bud, flower, early-fruit, mid-growth, and ripe, providing valuable data for pomegranate health monitoring. Tested on 587 images with 1109 labels, the model achieved recall of 0.888, precision of 0.916, mAP@0.5 of 0.943, and mAP@0.5:0.95 of 0.824. Despite challenges like small, hard-to-detect labels, the approach delivered satisfactory performance for accurate fruit identification.