The Journal of Science and Engineering Applications (JSEA)

Abstract

Rice, fed more than 50% of humans, rice is one of the most important plants in agriculture. Diseases can decrease the quantity and quality of this product, and crop losses from such diseases occasionally range between 30 to 60%. Plant diseases have always been one of the worst threatening situations for farmers. While the leaves of most plants have a unique disease characteristic, it still takes human eyes to spot them. This review also summarizes the outperformance of these approaches in solving various types of problems. This review discusses the constraints of existing studies and outlines prospective guidelines toward the significant enhancement of robust and well rice leaf disease detecting methods. Recent computer vision and Deep Learning developments introduced a disease classification system from leaf images, in modern agriculture. Identifying what type of disease the actual plant suffers from, a deep harvest cycle drives the world farmer to lose 37 percent of all the production due to non-accessibility of the details. (Source: ScienceDirect) The correct identification of disease for rice leaves is important as it can save on costs by enhancing the provision of noticed horticulture and managing the crop's future by maintaining a sound food-transferring system. This is done using a CNN model.