Optimizing rice plant disease detection with crossover boosted artificial hummingbird algorithm based AX-RetinaNet

Area of Science:

• Agricultural Science
• Computer Science
• Biotechnology

Background:

• Rice is a vital global food staple, but its yield is significantly impacted by diseases.
• Manual disease detection by farmers is time-consuming, resource-intensive, and prone to errors, leading to crop losses.
• Developing automated and accurate disease detection systems is crucial for sustainable rice cultivation and food security.

Purpose of the Study:

• To develop and evaluate a novel, automated approach for detecting and classifying rice plant diseases.
• To enhance the accuracy and efficiency of rice disease identification compared to traditional methods.
• To optimize the performance of a deep learning model using a metaheuristic algorithm for improved disease detection.

Main Methods:

• A new approach, crossover boosted artificial hummingbird algorithm based AX-RetinaNet (CAHA-AXRNet), was proposed for rice plant disease detection.
• The AX-RetinaNet model's hyperparameters were optimized using the crossover boosted artificial hummingbird algorithm (CAHA).
• Three datasets (rice plant, rice leaf, and rice disease) were utilized for training and validation to classify plants as healthy or unhealthy.

Main Results:

• The proposed CAHA-AXRNet approach demonstrated superior performance in rice plant disease detection and classification.
• The model achieved a high accuracy rate of 98.1% in distinguishing between healthy and diseased rice plants.
• Key performance metrics including precision, F1-score, specificity, and recall were employed to validate the model's effectiveness.

Conclusions:

• The CAHA-AXRNet approach offers a highly effective and accurate solution for automated rice plant disease detection.
• This novel method significantly outperforms existing approaches, promising to reduce crop losses and improve farming efficiency.
• The integration of advanced optimization algorithms with deep learning models shows great potential for agricultural disease management.