A novel face recognition method based on Local Zernike Moments

In this paper, an efficient face recognition scheme using Local Zernike Moments (LZM) is introduced. LZM is a localized version of Zernike Moments used successfully for character and fingerprint recognition. The superiority of LZM over LBP and Gabor methods on FERET dataset has been shown in previous studies. In this study, we demonstrate that Block Based Whitened Principal Component Analyses (BWPCA) can be successfully used with LZM. To increase the performance, we also determine and weight the face sub-regions used to create the feature vectors and the blocks used in dimensionality reduction step. The proposed method is evaluated on FERET dataset and it is shown that the obtained results are comparable to the best results in literature.