Smart Plant Disease Detection Using Deep Learning
Abstract
Plant diseases can damage crops and reduce both the quality and quantity of agricultural production. Finding a disease early is important because farmers can take action before it spreads. However, checking plant leaves by looking at them manually is not always easy, especially when different diseases have similar symptoms. This research presents a Smart Plant Disease Detection system that uses deep learning and image classification. The system is designed to identify tomato leaf conditions from images and classify them into three categories: Healthy, Early Blight, and Late Blight. A Convolutional Neural Network (CNN) was trained using a plant disease image dataset. The images were prepared and used to teach the model the visual patterns of different leaf conditions. The trained model was tested with a new image and successfully identified the condition of the leaf. In the project test, a healthy tomato leaf was predicted with 99.99% confidence. A simple Gradio interface was also created, allowing a user to upload a leaf image and get a prediction. The system can make plant disease identification quicker and easier and may help farmers and other agriculture-related users take early action..