Automated Detection of Tomato Plant Diseases: A Survey of Recent Advances

Abstract

Tomato farming is of great importance to global agriculture but is ever troubled by different plant diseases. Such plant diseases are detected earlier and accurately to allow intervention in a timely manner to salvage losses in yield. This survey paper serves the purpose of giving clear insights into the recent emerging technologies of machine learning (ML) and deep learning (DL) models in detecting the early tomato plant diseases. It analyses their use in various methodologies-the conventional image processing techniques, supervised ML algorithms and deep learning network architectures as well as the convolutional neural networks (CNNs) and their variants. The review discusses the pros and cons of the different approaches, identifies the common challenges encountered when put into real-world applications and suggests avenues for future research. This survey would serve as a one-stop, consolidated understanding of what is currently considered to be the state-of-the-art for automated tomato disease detection for research and practitioners looking forward to developing effective and powerful agricultural monitoring systems.