Model for Classifying and Predicting Fruit Leaf Diseases
Author Information
Author(s): Bai Chunhui, Zhang Lilian, Gao Lutao, Peng Lin, Li Peishan, Yang Linnan
Primary Institution: Yunnan Agricultural University
Hypothesis
Can the DINOV2-FCS model improve the classification and severity prediction of fruit leaf diseases?
Conclusion
The DINOV2-FCS model achieved high accuracy in classifying fruit leaf diseases and predicting their severity.
Supporting Evidence
- The model achieved 99.67% accuracy in disease classification.
- It reached an mIoU of 90.29% for semantic segmentation.
- The accuracy for disease severity classification was 95.68%.
- The model demonstrated strong generalization capabilities with an mIoU of 83.95% in testing.
Takeaway
This study created a smart model that helps farmers identify and understand how bad fruit leaf diseases are, making it easier to take care of their plants.
Methodology
The study used a large dataset of fruit leaf images and developed a model that combines advanced feature extraction techniques for classification and segmentation.
Potential Biases
Potential biases in the dataset could affect the model's generalizability.
Limitations
The model's performance may vary with different environmental conditions and species diversity.
Participant Demographics
Images were collected from various fruit diseases including apple, grape, and strawberry.
Statistical Information
P-Value
3.8e-4
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website