A Review of Machine Learning Approaches for Tomato Plant Disease Classification

Abstract

Tomatoes represent one of the most extensively cultivated and economically significant crops globally. Nevertheless, their productivity is frequently undermined by a range of plant diseases, which, if not promptly detected or effectively managed, can result in substantial yield reductions and economic repercussions. Traditional methods for disease identification are often labor-intensive, subjective, and reliant on expert knowledge, making them inefficient for large-scale agricultural operations. In recent years, machine learning (ML) has emerged as a promising avenue for automating and improving the accuracy of disease detection and classification. This review offers a comprehensive analysis of both conventional ML techniques and advanced deep learning (DL) methodologies specifically applied to the classification of tomato plant diseases. It examines the underlying methodologies of each approach, highlights commonly utilized datasets and evaluation metrics, and discusses the practical limitations associated with their implementation. Key challenges, such as data availability, model robustness, and the deployment of these solutions in real-world agricultural settings, are also addressed.