A Deep Learning-Based Method for Paddy Leaf Disease Detection and Growth Stage-Specific Treatment Recommendation.

Area of Science:

• Agricultural Science
• Computer Science
• Artificial Intelligence

Background:

• Paddy leaf diseases critically impact rice yield and quality, necessitating early detection and precise management strategies.
• Precision agriculture demands advanced tools for real-time monitoring and decision support in crop cultivation.

Purpose of the Study:

• To develop a deep learning-based decision support system for paddy leaf disease detection and growth stage prediction.
• To provide stage-specific treatment recommendations for identified diseases to enhance rice cultivation practices.

Main Methods:

• A dataset of paddy leaf images was curated and divided for training, validation, and testing.
• Lightweight Convolutional Neural Network (CNN) for growth stage prediction and transfer learning models (VGG16, ResNet50, InceptionV3, MobileNetV2) for disease classification.
• An ensemble method using average probability voting was employed to enhance classification accuracy and robustness.

Main Results:

• The ensemble model demonstrated superior accuracy, precision, recall, and F1-score compared to individual models.
• The system achieved improved robustness and generalization in paddy leaf disease detection and growth stage prediction.
• Experimental validation confirmed the effectiveness of the proposed deep learning framework.

Conclusions:

• The developed deep learning system offers an integrated solution for paddy disease detection, growth stage prediction, and treatment recommendations.
• This framework supports farmers and agricultural experts in making informed, timely decisions for effective disease management.
• The system contributes to advancing precision agriculture through AI-driven insights for sustainable rice production.