iMediPlants Recognizer An Intelligent tool to recognize Ayurvedic Medicinal Plants of Sri Lanka

Abstract

Sri Lanka is a country with a very old civilization and Ayurveda has been a widely practiced medical system throughout the ages. Herbal medicines are gaining popularity in Sri Lanka as they are safe to human health and affordable. However, due to unawareness of the value of the medicinal plants by people, who do not have an in-depth knowledge in identifying medicinal plants, and growing illegal trade and malpractices in the drug industry, precise identification of the medicinal plants is vital. iMediPlants is a standalone application for Sri Lankan medicinal plants which allows users to upload a digital image of a medicinal plant into the system and then the system will analyze the image and display the name of the medicinal plant. The whole idea behind the iMediPlants is to replace the manual process of identifying medicinal plants, by a computer aided identification system. This research aims at implementing such a system using up-to date image processing and artificial intelligence techniques. This research proposes a system for identifying Sri Lankan medicinal plants based on texture of the whole plant. Gray level co-occurrence matrix is used to extract texture features and neural network is used to classify medicinal plant species. The system was tested with twenty types of Sri Lankan medicinal plant species. The accuracy of identification using gray level co-occurrence matrix and neural network is 80% and indicated success in further research and enhancements.