Plant foliage Recognition based on Classification using Artificial Neural Network

Abstract

The state-of-the-art method to find the pictographic four types of foliage (flower, fruit, medical and tree) identification is proposed. Foliage is represented by a boundary of local feature using edge detection, followed by applying convex hull algorithm. In the second phase, ANN has been applied for simulating the system using the features identified in first phase. The proposed work resulted in an average identification rate of 96.75% and 94% with training and test data.