Neural network based Iris recognition system using Haralick features

Iris recognition has been paid more attentions due to its high reliability in personal identification recently. In this paper, an iris recognition system has been proposed. The steps of the proposed method include iris localization, normalization, feature extraction and matching of the iris pattern. To describe the iris data GLCM based Haralick features are used and for matching purpose probabilistic neural network is employed. Experiments are performed using iris images obtained from UBIRIS database. The method gives 97.00% correct classification rate.