What question did this study set out to answer?

The aim is to enhance rice leaf disease classification using a modified ResNet-50 model with advanced features.

Leveraging Squeeze Aggregation Excitation and Positional Encoding in ResNet‐50 for Rice Leaf Disease Classification

View Full Paper

Ask AI

Bookmark

Authors

Demonstrates improved rice leaf disease classification using a modified deep learning model, suggesting potential for better crop management.

The aim is to enhance rice leaf disease classification using a modified ResNet-50 model with advanced features.
Developed a modified model called ResNet-50+SaE-PE incorporating squeeze aggregation excitation and positional encoding.
Tested the model on a Rice Leaf Disease dataset with 4624 images, partitioned into training and testing sets.
Evaluated model performance using metrics such as accuracy, recall, specificity, precision, and F1-score.
Achieved an average accuracy of 99.89%, indicating high effectiveness.
Demonstrated superior performance compared to traditional identification methods and older models.
Identified Stochastic Gradient Descent as the most reliable optimizer for the task.

Bhuyan et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Demonstrates improved rice leaf disease classification using a modified deep learning model, suggesting potential for better crop management.