High Efficiency Iris Feature Extraction Based on 1-D Wavelet Transform

Abstract

In this paper, a new technique is proposed for high efficiency iris recognition, which adopts 1-D wavelet transform to extract iris texture feature and probabilistic neural network (PNN) to recognize iris pattern. A comparative experiment of existing methods for iris recognition is evaluated on CASIA iris image databases. The results reveal the proposed algorithm possesses a very high efficiency and low dimensionality of feature vectors than existing methods.