Memory efficient face recognition using Compressed Phase Component

Most of current algorithms for face recognition involve considerable amount of computations and hence they cannot be used on devices constrained with limited speed and memory. In this paper, we propose a novel solution for efficient face recognition problem for the systems that utilize small memory devices and demand fast performance. The new technique divides the face images into components and finds the discriminant phases of the Fourier transform of these components. The discriminant phases are found by the implementation of linear discriminant analysis (LDA) technique. Further, those discriminant phases are compressed to obtain a system that matches the criteria imposed by devices of limited memory and requiring fast recognition. The results show that in comparison with the use of spatial domain, compressed components frequency domain produces better and faster recognition performance for different range of compression ratios.