Iris Recognition System: A Novel Approach for Biometric Authentication

The main issue of this decade is security. Nowadays, there are different authentication systems based on a password, card and biometric are present. Password and card based systems are the failure due to humans' poor judgment. The researchers suggest that the biometric is most secure and prominent solution for systems security. Among the biometric systems, iris-based biometric systems are more secured. In this proposed system, segmentation technique utilized shape, intensity and localization of the pupil and iris. The segmented region is converted into the rectangular region by normalization process. The use of Gray Level Cooccurrence Matrix (GLCM) and Hausdorff Dimension (HD) extracts the texture based features. The extracted features were classified by supervised SVM machine learning algorithm. The performance of the proposed system shows the robustness of the system.