The Classification of Medicinal Plant Leaves Based on Multispectral and Texture Feature Using Machine Learning Approach

This study proposes the machine learning based classification of medical plant leaves. The total six varieties medicinal leaves-based dataset are collected from Department Agriculture, Islamia University Bahawalpur, Pakistan. These plants commonly named in English as (herbal) Tulsi, Peppermint, Bael, Lemon balm, Catnip, and Stevia scientifically Latin Ocimum sanctum, Mentha balsamea, Aegle marmelos, Melissa officinalis, Nepeta cataria, rebaudiana, respectively. multispectral digital image via a computer vision laboratory setup. For preprocessing step, we crop region leaf transform it into gray level format. Secondly, perform seed intensity-based edge/line detection utilizing Sobel filter draw five regions observations. A 65 fused features is extracted, being combination texture, run-length matrix, multi-spectral features. feature optimization process, employ chi-square selection approach select 14 optimized Finally, classifiers multi-layer perceptron, logit-boost, bagging, random forest, simple logistic deployed on an leaves dataset, observed that perceptron classifier shows relatively promising accuracy 99.01% compared to competition. distinct by 99.10% for 99.80% 98.40% 99.90% 99.20% Stevia.