Medicinal Leaf Identification using Machine Learning

Authors: Ravina Kolekar, Priyanka Ahire, Anushka Dholi

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: This project focuses on the development and implementation of a Convolutional Neural Network (CNN) model for the identification and classification of Ayurvedic plants. Ayurveda, an ancient medicinal system, relies heavily on diverse botanicals possessing therapeutic properties. Recognizing these plants accurately is crucial for their proper utilization in traditional medicine and scientific research.The proposed CNN framework leverages deep learning techniques to analyze and classify images of Ayurvedic plants. The model's architecture involves convolutional layers for feature extraction and classification layers for precise identification. Training the CNN involves a comprehensive dataset comprising high-resolution images of various plant species with annotations for supervised learning.Moreover, the project includes preprocessing techniques such as image augmentation and normalization to enhance the model's robustness and generalization capabilities. Evaluation metrics like accuracy, precision, recall, and F1 score are employed to assess the model's performance.The significance of this research lies in its potential to facilitate automated identification of Ayurvedic plants, aiding botanists, herbalists, and pharmacologists in authenticating plant species for medicinal purposes. Additionally, the project contributes to the convergence of traditional knowledge and modern technology, fostering interdisciplinary collaboration between botany and machine learning.

Keywords: Ayurveda Botanical classification Convolutional Neural Network (CNN) Deep learning Plant identification Traditional medicine Herbal remedies

Paper Id: 230432

Published On: 2024-01-08

Published In: Volume 12, Issue 1, January-February 2024

Cite This: Medicinal Leaf Identification using Machine Learning - Ravina Kolekar, Priyanka Ahire, Anushka Dholi - IJIRMPS Volume 12, Issue 1, January-February 2024.

Share this