A systematic literature review on emerging technologies, methodologies, dataset, challenges, and future trends for rice disease detection

Abstract Rice is a staple food crop cultivated worldwide and is essential to the livelihoods of millions. Among the common threats to rice cultivation, leaf diseases are particularly devastating, leading to reduced yields and significant financial losses. Traditional detection methods, relying on manual field inspections, are often tedious, time-consuming, and require specialized expertise. In recent years, automated rice disease detection technologies have seen significant advancements. This paper presents a review of emerging techniques for rice disease detection, categorizing prior studies based on artificial intelligence (AI), image processing (IP), and the Internet of Things (IoT). It also examines the types of rice diseases, available datasets, hyperparameter optimization strategies, and their influence on model performance highlighting the importance of fine-tuning. The review is based on 125 high-quality publications and identifies promising future research directions by addressing existing gaps at the intersection of AI, IP, IoT, and rice disease detection.