| 作成者 |
|
|
|
|
|
| 本文言語 |
|
| 出版者 |
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 開始ページ |
|
| 終了ページ |
|
| 会議情報 |
|
| 出版タイプ |
|
| アクセス権 |
|
| 権利関係 |
|
| 権利関係 |
|
| 概要 |
Ensuring varietal purity in rice production is vital, but currently relies on labor-intensive, subjective, and hard-to-scale methods. This study introduces an AI-based Rice Seed Varietal Purity Testin...g Machine that classifies NSIC Rc160, Rc216, and Rc222 seeds using a 48MP CMOS camera paired with a custom-trained YOLOv11s-seg deep learning model featuring segmentation. The system achieved a mean Average Precision (mAP@0.5) of 96.9% and an F1 score of 93%, with classification accuracies of 96.94% for Rc160, 96.38% for Rc216, and 95.83% for Rc222. Its processing speed is on par with manual testing while meeting accuracy thresholds for reliable use. Real-time seed sorting is managed by an ESP32-driven electromechanical system. Despite Rc222’s morphological complexity, this solution offers a scalable, intelligent, and efficient approach for varietal certification, addressing key challenges in Philippine rice agriculture.続きを見る
|
Mendeley出力
2025_p0576