A Novel Biometric Authentication using Contourlet Transform and Enhanced MSVM
Identifying a person using biometrics is increasingly becoming popular these days. While using biometrics, a number of modalities are available in which face recognition, fingerprint and palmprint identification are widely used. Among them, palmprint is the widely accepted solution because of its richness of features and the low cost involved in acquisition. In this paper, a novel biometric authentication scheme that uses Contourlet transform for performing analysis has been proposed. This transform is a pyramidal directional filter bank which can be applied to capture both local and global features of palmprint. In this process, a set of features are extracted first. Now, dimensionality reduction is carried out by calculating the energies for each subband and then uses them to minimize the vast amount of coefficients generated.