DEVELOPMENT OF A PEST AND DISEASE DETECTION SYSTEM USING SELECTED MACHINE LEARNING ALGORITHM

This project focuses on developing a pest and disease delection system using machine learning to enhance agricultural productivity and crop heath. Traditional pest and disease identification methods are time-consuming, require expert knowledge, and are ofen inaccurate under field conditions.By automatingdetection through machine learning, this project aims to provide faster,more reliable identification of threats to crops,thereby supporting timely intervention and reducing losses. The project follows the Agile methodology, emphasizing iterative development. A diverse dataset of plant images exhibiting various pests and diseases was collected and labeled. Machine learning models were trained to detect and classify these conditions, and a user-friendly interface was built to support farmers and agricultural professionals in real-time decision-making. Continuous testing and refinement were performed throughout development to improve accuracy and usability. Project results demonstrate significantly faster detection with high precision across multiple crop types. The system proved effective in real-time field tests,reducing manual inspection time by up to 125% while maintaining a high detection accuracy. In conclusion, the project shows the potential of machine learning to transform pest and disease management in agriculture. It lays a strong foundation for future enhancements,including expanding the dataset and integrating with IoT-based monitoring systems.