Advancements in maize disease detection: A comprehensive review of convolutional neural networks

This review article provides a comprehensive examination of the state-of-the-art in maize disease detection leveraging Convolutional Neural Networks (CNNs). Beginning with the intrinsic significance of plants and the pivotal role of maize in global agriculture, the increasing importance of detecting and mitigating maize diseases for ensuring food security is explored. The transformative potential of artificial intelligence, particularly CNNs, in automating the identification and diagnosis of maize diseases is investigated. Various aspects of the existing research landscape, including data sources, datasets, and the diversity of maize diseases covered, are scrutinized. A detailed analysis of data preprocessing strategies and data collection zones is conducted to add depth to the understanding of the field. The spectrum of algorithms and models employed in maize disease detection is comprehensively outlined, shedding light on their unique contributions and performance outcomes. The role of hyperparameter optimization techniques in refining model performance is explored across multiple studies. Performance metrics such as accuracy, precision, recall, F1 score, IoU, and mAP are systematically presented, offering insights into the efficacy of different CNN-based approaches. Challenges faced in maize disease detection are critically examined, emerging opportunities are identified, and future research directions are outlined. The review concludes by emphasizing the transformative impact of CNNs in revolutionizing maize disease detection while highlighting the need for ongoing research to address existing challenges and unlock the full potential of this technology. • Comprehensive review of CNN applications in maize disease detection. • Novel preprocessing strategies enhance model performance and accuracy. • Expanded hyperparameter optimization techniques for improved CNN training. • Inclusion of recent studies showcasing advancements in deep learning methods. • Insights into future research directions for sustainable agriculture practices.