AI-Based disease detection system for rice crop

Abstract

The rice is priority food in Rwanda, With an average productivity of 5.8 t/ Ha. Rice is grown over 12,400Ha of marshlands in two seasons which makes around 80,000 MT per year. Although there has been a rapid rise in rice production compares to the past decade the country has not yet achieved self-sufficiency (jICA Magazine, 2023) , but not only in Rwanda the rice productivity in all over the world fluctuates significantly from region to region due to various factors such as pest and diseases, soil type, soil fertility, rainfall pattern, flood, drought, water logging and climatic condition (Bin Rahman, 2023),This research proposes the development of an AI-based disease detection system for rice crops, aiming to address the persistent challenges faced by farmers in detecting and managing diseases effectively. Traditional methods reliant on manual observation lack accuracy and efficiency, leading to significant crop losses. Leveraging advanced machine learning techniques and Internet of Things (IoT) technologies, the proposed system offers a precise, scalable, and user-friendly solution. Through the integration of convolutional neural networks (CNNs) and IoT sensors, the system aims to achieve early and accurate detection of various rice diseases. The study encompasses the development, implementation, and evaluation of the system's performance, with the goal of revolutionizing rice farming practices and contributing to food security and economic sustainability in rice-growing regions.