Medicinal plant leaf classification using gradient-based features

Abstract

Herbs are used since ancient times for healing various ailments in traditional medicine in various parts of the world. The identification or recognition of such plants has become a challenge for many, and the proposed method will help in overcoming these issues through classification of leafs using gradient-based features and Nearest Neighbour classifier in Support Vector Machines (SVM). The pre-processing steps involve binarization using Otsu’s method, noise removal using median filter and resizing. After preprocessing, the preprocessed leaf image will be extracted and the gradient-based features will be analyzed using