Plant Leaf Disease Identification Based on Machine and Deep Learning Methods

The substantial effects of plant diseases on agriculture, which result in losses in production of crops, financial losses, and food insecurity, highlight the necessity for plant leaf disease detection and categorization. In order to effectively control illnesses, stop them from spreading to other plants, and develop tailored treatment plans, early and precise disease identification is essential. Various strategies are used for plant leaf disease diagnosis, such as manual symptom assessment, computer image processing, and visual examination by skilled professionals. Traditional methods rely on human expertise to visually identify disease symptoms based on characteristic patterns, discolorations, or lesions on plant leaves. These techniques, however, are laborintensive, time-consuming, and sensitive to bias and mistakes. The development of automated and computer-based techniques for the identification of plant leaf disease through the use of ML (machine learning) algorithms and digital imaging technology has garnered increasing attention in recent years. There are still a number of issues with the DL (deep learning) and ML algorithms. In order to identify research gaps and address them in the future while developing a unique framework, this study presents an examination of ML and DL models.