Plant Recognition using Hog and Artificial Neural Network

Abstract

This paper presents a plant leaf recognition system being implemented through Artificial Neural Networks. The system proposed is designed using MATLAB Software which takes a leaf image from the user and classifies, recognizes the plant species and shows all the relevant details about the plant.it also incorporates a webpage from various plant databases. The leaf features are extracted by using a HOG (Histograms of Oriented Gradients) vector and the ANN(Artificial Neural Network) is used in training through Backpropagation. We have extracted the HOG features from the flavia datasheet of leaves and trained them in the Neural Network. The results were nearly perfect and the accuracy of the program implemented is very high compared with other models.