Early Detection of Fungal Diseases in Tomatoes using Convolutional Neural Networks

Background: Tomatoes play a pivotal role in global agriculture fostering economic growth while also significantly enhancing food security. This paper introduces a data-based approach to improve the early detection of fungal infections in tomatoes. Methods: The study utilises a dataset comprised of 4,362 high-resolution images obtained from Mendeley, showing healthy tomato plants alongside those damaged by leaf mold, early blight and late blight diseases. Deep learning techniques have been applied to develop a Convolutional Neural Network (CNN) for disease classification based on these images. Result: The CNN model exhibits an overall accuracy rate of 90.83%, underscoring its efficacy in identifying fungal growth in tomato plants. The study proposes the use of machine learning in the detection and treatment of fungal diseases affecting tomatoes, which consequently leads to increased crop yield and quality preservation. It recommends the development of automated tools for farmers to detect and respond to disease outbreaks, thus improving their agricultural practices.