An Advanced Review of Machine Learning Methods for Identifying Medicinal Plant Leaf Diseases

Abstract

The identification and classification of medicinal plant leaf diseases using machine learning techniques have become essential in agricultural and pharmaceutical research. This paper extends previous studies by reviewing advanced machine learning classifiers—XGBoost, Naïve Bayes (NB), Logistic Regression (LR), and k-Nearest Neighbors (KNN). Using the medicinal plant dataset available at [https://data.mendeley.com/datasets/hb74ynkjcn/5], we analyze classification performance, computational efficiency, and practical applicability. Unlike earlier studies that focused on Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF), this review highlights new approaches, including boosting methods and probabilistic classifiers.

Plant phenotyping relevance

薬用植物の葉病害を機械学習で識別・分類する手法をレビューし、複数分類器の性能と計算効率を比較しているため、植物病害状態のフェノタイピング手法が中心です。

Code and data availability

The paper's medicinal plant leaf disease classification experiments use a public Mendeley Data leaf image dataset (hb74ynkjcn/5), explicitly cited as the dataset analyzed in the study.