A review on machine learning and deep learning image-based plant disease classification for industrial farming systems

Abstract

Plants can be affected by various diseases. As such, the early detection of crop diseases plays an essential role in the farming industry. However, such detection requires extensive pathogen knowledge and is costly and labor-intensive. These challenges present an attractive opportunity to leverage machine learning (ML) and deep learning (DL) techniques to automate the detection of crop diseases. From this perspective, we present a review paper that showcases image-based plant disease detection and classification systems and discusses success stories using ML and DL techniques. In this review paper, we examine various aspects of these systems, including the sources of plant datasets, algorithm types and techniques used in ML and DL. The findings of this review paper inspire future research by highlighting the potential challenges in applying ML and DL to plant disease and pest detection. Additionally, it proposes potential solutions to overcome these challenges, paving the way for further advancements in developing and implementing automated systems for plant disease detection and classification. This work serves as a valuable resource for researchers and practitioners in the field, guiding their efforts toward more effective and accessible solutions for crop disease management.