Artificial Intelligence-Based Approaches for Plant Leaf Disease Detection

Abstract

Medicinal plants have long been central to traditional healing systems, especially in Ayurveda, which promotes balance between body, mind, and environment. Accurate identification of these plants is essential for safe use, yet conventional methods often rely on subjective human judgment. This study explores how artificial intelligence can support plant recognition by using machine learning (ML) and deep learning (DL) techniques to classify medicinal leaves. A diverse image dataset was analyzed based on visual features such as color, texture, shape, and size. ML models—including Support Vector Machines, Decision Trees, and Random Forests—were compared with advanced DL architectures like VGG16, InceptionV3, and Vision Transformers. To improve model generalization and reduce overfitting, data augmentation and transfer learning were applied. Results showed that DL models consistently outperformed ML approaches, with the transformer-based model achieving the highest classification accuracy. These findings suggest that AI-powered systems can make medicinal plant identification more reliable and scalable. The study offers practical insights for digital health applications and contributes to the modernization of traditional medicine through technology-driven solutions.