AI-Powered IoT Framework for Early Disease Detection in Swarna Rice: A CNN-Based Approach to Precision Agriculture

Abstract

Rice (Oryza sativa L.) cultivation is crucial to global food security but is highly susceptible to diseases like blast, bacterial leaf blight, and brown spot, which significantly reduce yields. Traditional disease detection methods rely on manual inspection, which is time-consuming, error-prone, and often delays timely interventions. To overcome these challenges, we introduce AgriCapture, an intelligent IoT-ML-based rice disease detection and monitoring system. The system employs a Convolu- tional Neural Network (CNN) for real-time disease classification, alongside IoT sensors for continuous environmental monitoring, providing farmers with valuable, actionable insights. A Streamlit- based web application allows users to upload images of rice leaves for instant disease identification, while simultaneously displaying real-time data on temperature, humidity, and soil moisture via Firebase. Trained on a curated dataset, the CNN model achieves 87.5% accuracy, ensuring reliable disease classification. By seamlessly integrating automated disease detection with real- time monitoring, AgriCapture offers a cost-effective, scalable, and user-friendly solution to mitigate rice crop losses and improve agricultural productivity. This system is not only adaptable to small-scale farmers but also has the potential to scale globally, enhancing sustainable agriculture and global food security.