A comparative analysis of Zernike moments and principal component analysis as feature extractors for face recognition

Abstract

This paper describes the comparison between Principal component analysis (PCA) also known as eigenfaces and Zernike moments (ZM) as feature extractors, used in face recognition. These feature extraction methods are still being research until today even though the techniques may either be hybrid or fusion. The study look into the capability of these two feature extraction methods to recognize face due to changes in illumination condition, pose, facial expression and others. The experiment carried out utilizes the earliest eigenfaces technique adopted from Turk and Pentland [1] and ZM polynomials [2]. The classification technique employed in the recognition stage is a simple Euclidean square distance classifier or nearest neighbor (NN). The experiments utilized database face images from Olivetti research laboratory (ORL) consisting of 40 subjects of 10 images each where none of them are identical [3]. They vary in position, rotation, scale and expression, with and without glasses. From the comparative study, the outstanding feature extraction method is considered for face recognition system.