Related Experiment Videos
Metaheuristic optimization based improved neural network for the timely prediction of paddy leaf diseases
1Department of Electronics and Communication Engineering, Dr. G. U. Pope College of Engineering, Sawyerpuam, Tamil Nadu, 628251, India.
Summary
This study introduces an AI system for early detection of rice leaf diseases using image analysis. It enhances crop yield and reduces environmental impact through optimized fertilizer recommendations.
Area of Science:
- Agricultural Science
- Computer Science
- Plant Pathology
Background:
- Rice is a crucial global food staple, but its productivity is threatened by leaf diseases.
- Effective disease prediction and management are essential for food security and economic stability.
Purpose of the Study:
- To develop an advanced image analysis system for prompt and accurate detection of paddy leaf diseases.
- To integrate disease prediction with precision agriculture for optimized fertilizer management.
Main Methods:
- Image pre-processing using Adaptive Gabor Filter (AGF) and feature extraction with Histogram of Oriented Gradient (HOG).
- Disease classification using Optimized Capsule Networks (CapsNet) optimized with Glow Worm Swarm Algorithm (GWSA).
- Implementation of a Fertilizer-Based Disease Management (FBDM) approach for site-specific nutrient recommendations.
Main Results:
- The proposed system accurately identifies diseases affecting paddy foliage.
- Optimized CapsNet demonstrates improved generalization and robustness in disease detection.
- The FBDM approach enables precise fertilization, enhancing plant resistance and reducing environmental impact.
Conclusions:
- The integrated system offers a robust solution for early rice disease detection and management.
- Precision agriculture techniques, guided by AI, can significantly improve crop yields and sustainability.
- This approach supports efficient resource utilization and minimizes environmental pollution in rice cultivation.