Base and apex angles and margin types-based identification and classification from medicinal plants' leaves images
by Suvarna S. Nandyal; Basavaraj S. Anami; A. Govardhan
International Journal of Computational Vision and Robotics (IJCVR), Vol. 3, No. 3, 2013

Abstract: In this paper, we have proposed a method for medicinal plants identification and classification from their leaves images based on base angle, apex angle and margin type. In Ayurveda, medicines are extracted from leaves, roots and stems. Many a times, the plants are recognised by their leaves. Therefore, we have considered leaves images in this work. Adoptive Otsu threshold method is used for segmentation of leafy part from the background. The database includes altogether 900 leaves images representing 18 shapes with different base and apex angles and margins. The leaf images are classified based on features such as widest part of leaf and its location, stem angle, base angle and margin. The two angles and margin features are extracted using trigonometric and chain code approaches respectively. The proposed approach is robust, position, scale and rotation invariant and works for even tender leaves. The average recognition and classification accuracy of 98% is obtained. The work assists novice users of Ayurvedic practitioners.

Online publication date: Fri, 18-Jul-2014

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Vision and Robotics (IJCVR):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com