A deep learning optimized model for classification and detection of rice leaf diseases.

Area of Science:

• Agricultural Science
• Computer Science
• Artificial Intelligence

Background:

• Plant diseases, particularly rice diseases, pose a significant threat to global food security by impacting food productivity, quantity, and quality.
• Accurate and timely diagnosis of rice leaf diseases is essential for effective disease management and mitigation strategies.

Purpose of the Study:

• To develop and evaluate a deep learning model for the accurate categorization and forecasting of rice plant diseases.
• To enhance the performance of deep learning models for plant disease classification using an optimization algorithm.

Main Methods:

• The study employed a deep learning framework integrating a deep maxout network (DMN) and a deep autoencoder (DAE).
• The remora optimization algorithm (ROA) was utilized to optimize the learning parameters of the deep model, aiming for improved convergence and avoidance of local minima.
• The proposed ROA-DM method was applied to a rice leaf dataset for disease detection and classification.

Main Results:

• The ROA-DM method demonstrated high accuracy and precision across various rice disease categories.
• Experimental results, including confusion matrices, showed strong training and validation performance.
• The optimized learning approach achieved a classification accuracy of 98.5% for rice leaf diseases.

Conclusions:

• The developed ROA-DM deep learning model shows significant potential for accurate and precise identification of rice leaf diseases.
• The integration of the remora optimization algorithm enhances the performance of deep learning models in plant disease classification.
• This approach contributes to improving disease management practices, thereby supporting food security.