Feature Extraction Using Discrimination Power Analysis with Palmprint in Biometric System

Abstract

There are many methods used to distinguish persons from each other using palmprints, fingerprints, giats, hand geometry, sound waves, shape of the ear, the color of the eyes and the iris, …etc. Discrimination power analysis (DPA) is powerful to extract proper feature for palmprint, after applying (DPA) on the true color image of palmprint, some of the features and coefficients are choosing to construct vectors including these value of features. DPA is a statistical analysis, based on the image coefficients properties after sequence of process on the true color image. The discrimination power of all the coefficients is not the same and some of them are discriminant than others, the higher true of recognition rate depend on feature vectors of Discriminant Coefficient (DCs). It searches for the coefficients which have large power to discriminator different classes better than other, the performance based on (DPA) and selected (DCs) is better with less complexity.