Classification and Recognition of Herbal Leaf Using SVM Algorithm.

Ayurveda is an ancient holistic healing system, which is nearly 5000 years old that uses natural herbal leaves to heal various diseases. Herbal leaves have been widely utilized by the industries like food, medical and cosmetic. But most of the herbal plants are endangered. So, there is a need to identify and regrow them for the use of future generations. So it is very indispensable to develop an automatic system to identify species correctly. This paper proposes an algorithm for recognition of herb leaf through image processing and classification technique using Support Vector Machine (SVM) classifier. The scale invariant feature transform (SIFT) method is used for the feature extraction in leaves. The work is carried out using 5 species of leaves in MATLAB environment. SVM classifier is trained with 273 samples and tested with 129 samples of herbal and non-herbal leaves are presented in this paper. The output showed that the system reached the recognition rate with 92.25% accuracy. The system is convenient to use and cost effective for the botanist, Scientist and Researcher to identify herbal leaves.