An Efficient Human Identification Through Iris Recognition System

As a part of a growing information society, nowadays the issue of security is more crucial than ever. In order to achieve high level of security, the potential of accurately recognize subjects based on their unique measurable physiological or behavioral characteristics has been receiving an increased concern by the research and development community. As biometrics has advanced, iris has been considered a preferred trait because unique pattern texture, lifetime stability, and regular shape contribute to good segmentation and recognition performance. The incredible uniqueness of iris patterns as well as the ability to capture iris images non-invasively has motivated us to develop automated system for iris recognition based on 2-D iris images. The 2DPCA (two-dimensional Principal Component Analysis) and GA (Genetic Algorithm) have been used as feature extraction and feature selection techniques for reducing the dimensionality of iris features without the loss of relevant Information. The Back Propagation Neural Network (BPNN) is implemented using Levenberg–Marquardt’s learning rule for iris recognition. The experimental results illustrated that the 2DPCA-GA achieved a high classification accuracy of 96.40 %.