COMBINATION OF PCA AND SVD WITH DISCRETE CURVELET ALGORITHM IN IRIS RECOGNITION

Abstract

The Iris texture is one of the key factors in iris recognition. Although Curvelet transform is being widely used to recognize human’s iris, its complex coefficients create large featured dimensions of images. This paper proposes an iris recognition method based on Curvelet, Principal Component Analysis (PCA) and Singular Value Decomposition (SVD). Experimental results showed that the iris recognition method using the first layer Curvelet coefficients improved the iris recognition rate. In addition to that, this method also reduces featured dimensions and improves the recognition speed.