Improving Facial Image Recognition based Neutrosophy and DWT Using Fully Center Symmetric Dual Cross Pattern

Abstract

Face recognition is one of the most commonly used biometric features in the identification of people. In this article, a novel facial image recognition architecture is proposed with a novel image descriptor which is called as fully center symmetric dual cross pattern (FCSDCP) The proposed architecture consists of preprocessing, feature extraction and classification phases. In the preprocessing phase, discrete wavelet transform (DWT) and Neutrosophy are used together to calculate coefficients of the face images. The proposed FCSDCP extracts features. LDA, QDA, SVM and KNN are utilized as classifiers. 4 datasets were chosen to obtain experiments and the results of the proposed method were compared to other state of art image descriptor based methods and the results clearly shows that the proposed method is a successful method for face classification. © 2019 MECS.