Deep Learning and Paddy Leaf Detection: A Comprehensive Analysis

The agricultural sector plays a crucial part in ensuring security of food supply worldwide, and rice cultivation stands as a cornerstone for many economies worldwide. Over recent decades, the emergence of novel plant diseases has become a recurring challenge, significantly impacting crop development, particularly due to farmers' unawareness of these new diseases. Prompt identification of plant diseases is paramount for effective crop health monitoring, and the transition from manual monitoring to automated plant monitoring technologies is essential for farmers. Early identification of crop diseases not only facilitates timely intervention but also enhances agricultural productivity. Among agricultural products, paddy crops hold particular significance as a primary and essential food source. Leaf diseases have been identified as a major detriment to paddy crop productivity and quality. Addressing this issue promptly is imperative to minimize the resulting damage. In response to the need for accurate and rapid disease diagnosis, researchers have made substantial strides in automating the detection process, employing advanced computational approaches. This proposed study aims to investigate innovative techniques and strategies within the domains of deep learning, machine learning, and optimization to effectively identify and categorize paddy leaf diseases. This study evaluates the merits and demerits of current methods, along with the datasets they utilize, from the years 2020 to 2024. The findings from this study offer a comprehensive overview of the current landscape, providing valuable insights for future research directions. The identified gaps underscore the necessity for further exploration and innovation, particularly in developing models capable of adapting to diverse environmental conditions and addressing unforeseen challenges in real-world agricultural settings.