Pioneering disease prediction in cinnamon leaves using machine learning: a systematic literature review

Abstract

The integration of Machine Learning (ML) in agricultural disease prediction has become increasingly prominent. This review paper explores the evolution of techniques used for predicting diseases in cinnamon leaves and analyzes common cinnamon leaf diseases, drawing on research conducted up to 2023. The paper highlights the evolution of ML methodologies, particularly in the areas of image processing, feature extraction, and classification algorithms. It provides an in-depth analysis of various approaches, such as Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and Random Forests, evaluating their effectiveness in disease prediction. From an initial set of 100 studies, 22 were selected for detailed analysis based on their relevance and contribution to the field. Additionally, the review addresses the challenges associated with developing reliable ML models. Through the synthesis of findings from multiple studies, this paper offers a comprehensive overview of current research in cinnamon leaf disease and prediction, identifying existing gaps and proposing directions for future investigations to improve the precision and applicability of ML driven solutions in agriculture.