A new face recognition method for corrupted images

Corrupted face image is one of the important obstacles that machine vision systems face when trying to recognize faces. In this paper, we propose a new solution that deals efficiently with this significant problem when face recognition application is employed. The new technique applies the principal component analysis to the phase spectrum of the Fourier transform of the covariance matrix constructed from the MPEG-7 Fourier Feature Descriptor vectors of the images. It will be shown that the proposed technique increases the face recognition rate when applied to images of low resolution and corrupted by noise, compared to other known systems. Experiments on the ORL face database are reported to demonstrate the effectiveness of the proposed technique.