<![CDATA[Agriculture, particularly rice cultivation, is crucial for Indonesia's food security; however, production is often hindered by pests and diseases. With over 30 million hectares of rice fields and millions of farmers relying on this staple crop, the impact of these challenges is significant, threatening both livelihoods and national food supply. This study aims to develop a rice plant disease diagnosis system using Convolutional Neural Network (CNN) methods implemented in a Streamlit-based application. Data were obtained from an open dataset on Kaggle, which includes images of healthy and infected rice leaves. The Streamlit application facilitates users in uploading images and receiving real-time diagnoses. Results show that the CNN model achieved an accuracy of 96.03% in identifying diseases, demonstrating a strong ability to recognize patterns in leaf images. This system offers an efficient solution to help farmers quickly and accurately detect rice diseases, contributing to increased agricultural productivity and food security in Indonesia]]>
Copyrights © 2025
