Medicinal plants segmentation using thresholding and edge based techniques

Abstract

Medicinal value is the integral essence and active ingredients in plants. Knowledge about medicinal plants is more essential to mankind. Manual identification of medicinal plant requires prior information and needs the assistance of botanist because all the plants have similar color, shape and texture characteristics and it is time consuming process. Due to this reason it is very much essential for Pharmacists, Botanist and Ayurveda Practitioners to know how to identify medicinal plants through computer technologies. Image processing technology is used for identification and classification of medicinal plants. The following three general phases of image processing techniques are applied to all types of images namely pre-processing, segmentation and feature extraction. Among all the stages of image processing, segmentation phase partition an image into number of regions and segment the object from background. It improve the representation of an image into more meaningful and easier to analyze the features. The medicinal plants segmentation is carried out in this proposed work. The common and famous segmentation techniques like edge detection and thresholding methods are applied on different shapes of flowers, leaves and fruits/seeds images. This paper explores the comparisons of above two methods and effects of filtering and enhancement techniques.