Iris Recognition using Gray Level Co-occurrence Matrix and Hausdorff Dimension

Abstract

Biometrics authentication is the only accurate solution for personal identification and security problems. Password incorrect use and misapplication, intentional and inadvertent is a gaping hole in security. These results are mainly occurs due to Poor human judgment, carelessness and due to tactlessness. Biometric removes all these types of security mistakes. In iris recognition system identification and verification is one of the efficient method. The objective of this proposed system is to analyze the performance of iris. The segmentation of the iris utilizes shape, intensity, and location information for pupil or iris localization and performs normalization of the iris region by unwrapping the circular region into a rectangular region. The feature extraction of iris was done by biometrics GLCM (Gray Scale Co-occurrence Matrix) and HD (Hausdorff Dimension. The BGM (Biometric Graph Matching) algorithm is used, which is used to match the graph between the training image and test image of the iris biometric. The BGM algorithm uses graph topology to define different feature values of the iris templates. A SVM (Support Vector Machine) classifier is used to distinguish between genuine and imposter. The results give better performance and authentication than the existing method.