AN EMBEDDED SYSTEM FOR IDENTIFICATION AND CONFIRMATION OF AYURVEDIC PLANT USING KNOWN LEAF IMAGE DATABASE

Nithu V.C., Liju Philip, Deepa J.

Abstract: In Ayurveda, medicines are extracted from leaves, roots and stems of medicinal plants. Many a times, the plants are recognised by their leaves. Recognition of the medicinal plants from forests, fields, farms etc. is very tedious and time consuming. This paper presents the design and development of an embedded system to identify ayurvedic plants from their leaf images based on the geometric features. The system is developed based on how human beings recognize leaves in the real world. The database of known ayurvedic plants was created for the purpose which includes features of leaves extracted from twenty types of plants that have high medicinal value. The algorithms of image processing in OpenCV are used for feature extraction from leaf image. The system has been implemented on Raspberry Pi. Experiments conducted on a total of 200 leaves, 10 numbers of leaves collected from each of twenty different plants shows an average 96% accurate identification of plants from the features of their leaves

Keywords: Ayurvedic Plants, Segmentation, Feature Extraction, Image Acquisition, Medicinal Plants

DOI: https://doi.org/10.15623/ijret.2017.0602024