Face Recognition Based on Local Statistical Features and Artificial Neural Network

Abstract

In this paper a face recognition method based on image segmentation, statistical features, and neural network is proposed which is composed of three main steps of (1) preprocessing, (2) extraction of statistical features including mean, standard deviation, skewness, and kurtosis, and (3) classification using a perceptron neural network with one hidden layer. The proposed method benefits the advantage of simplicity in implementation. In addition, the simulation results show that the proposed method could achieve the recognition accuracy of 99.8% which outperforms the competitive methods of principal component analysis (PCA) (8.25-13.34% improvement), k-nearest neighbors (11.95-17.54% improvement), local binary pattern (4.45-10.04% improvement), support vector machine (SVM) combined with the PCA (0.19-2.18% improvement), and convolutional neural network (up to 0.64% improvement).