A review on Machine Learning Approaches for Plant Disease Diagnosis

In recent years, the identification and classification of plant and crop diseases have significantly increased due to the use of machine learning (ML) algorithms. This rise is attributed to the remarkable precision and potential of ML methods in automating plant disease identification. This comprehensive analysis examines various ML approaches and image processing techniques used to identify and classify diseases such as false smut, bacterial leaf blight, brown leaf spot, leaf scald, rice blast, rice tungro, sheath blight, and stem rot also we embed the rice crop major disease, the disease caused by, symptoms of diseases, factors that trigger the disease and in what stage the disease affect the crops. Here we register the common issues in generalization of models, crop disease diagnosis process stages. We explore advanced ML techniques as well as conventional ML methods like random forest (RF), naive bayes (NB), decision tree (DT), k-nearest neighbors (KNN) and support vector machine (SVM). A tabular summary of the results of these methods is provided in this paper. This article also discusses about the challenges and limitations associated with applying ML methods to detect plant disease, offering insights into the constraints practitioners face. Our goal is to provide a thorough overview of the latest developments in machine learning algorithms for crop disease diagnostics by integrating existing research. Ultimately, this study enhances understanding of the diseases, methodologies, challenges, and future directions for using ML in sustainable agriculture.