PM-YOLO: A Model for Detecting Powdery Mildew in Rubber Trees
Author Information
Author(s): Li Yuheng, Chen Qian, Zhu Jiazheng, Li Zengping, Wang Meng, Zhang Yu
Primary Institution: Hainan University
Hypothesis
Can a deep-learning-based model effectively detect and grade powdery mildew on rubber trees?
Conclusion
The PM-YOLO model significantly improves the detection accuracy of powdery mildew on rubber trees, outperforming existing methods.
Supporting Evidence
- PM-YOLO achieved 86.9% mean average precision and 85.6% recall.
- The model outperformed YOLOv10 by 7.6% in mAP and 8.2% in recall.
- A dataset of 6200 images with 38,000 annotations was created for training.
- The model integrates advanced modules for better detection of small targets.
Takeaway
This study created a smart tool that helps farmers find and measure a disease on rubber tree leaves, making it easier to take care of the trees.
Methodology
The study used a deep-learning model called PM-YOLO, which was trained on a dataset of 6200 images of rubber tree leaves with 38,000 annotations.
Limitations
The dataset may not cover all environmental conditions and challenges like occluded leaves or poor lighting.
Statistical Information
P-Value
0.0001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website