Principal Valued Complex Spectrum Energy Ensembles for Palmprint Representation

The palmlines in palmprint exhibit different line widths. These linewidths are responsible for different kinds of spectral components. In this work, the energies of the principal valued complex spectrum are employed to characterize the various spectral components of the palmprint. The proposed work attains a discrimination by employing the energies of improved the spectral coefficients. From experimental results on polyU Ver. II database, it is evident that the proposed approach demonstrates significant improvement as compared to state-of-the-art approaches.