Iris Recognition: An Emerging Biometric Technology


Iris recognition

Iris Recognition System

Iris Recognition Based on Neural Networks

Phase-Based Iris Identification

Moving Average Filter Iris Recognition

Iris Identification Based on 2D Wavelet

DCT-Based Iris Recognition

Hybrid Iris Recognition

LBP Iris Recognition

Iris Recognition Based on Genetic Algorithms

One-to-One Iris Recognition System

External resources

Advanced Source Code .Com

Neural Networks .It

Genetic Algorithms .It

Face Recognition .It

DCT-Based Iris Recognition

Download now Matlab source code
Requirements: Matlab, Matlab Image Processing Toolbox.

Pattern recognition methods can be classified into semantic and nonsemantic approaches. The use of the Karhunen-Loeve Transform (KLT) for object recognition and, in particular, face recognition, are examples of nonsemantic techniques. The advantage of such methods arises from the automatic generation of suitable feature vectors by the KLT. Advanced feature extraction techniques find extensive use in the increasingly important domain of biometric identity authentication. As security becomes an issue of importance, biometrics and iris recognition in particular are attracting great interest. The human iris, a thin circular diaphragm lying between the cornea and the lens, has an intricate structure with many minute characteristics such as furrows, freckles, crypts, and coronas. For every subject, these characteristics are unique as a result of the individual differences that arise in the development of anatomical structures during embryonic development. Apart from general textural appearance and color, the finely detailed structure of an iris is not genetically determined but develops by a random process. The iris patterns of the two eyes of an individual or those of identical twins are completely independent and uncorrelated. Additionally, the iris is highly stable over a personís lifetime and lends itself to noninvasive identification because it is an externally visible internal organ. Pioneering work on iris recognition was done by Daugman using Gabor wavelets.

We have developed a method for iris matching using zero crossings of a Discrete Cosine Transform (DCT) as a means of feature extraction for later classification. The DCT of a series of averaged overlapping patches are taken from normalized iris images and a small subset of coefficients is used to form subfeature vectors. Iris codes are generated as a sequence of many such subfeatures, and classification is carried out using a weighted Hamming distance metric.

We have compared our results with a publicly available system for iris recognition developed by Libor Masek and Peter Kovesi available here. Our code has been tested with CASIA Iris Databaseachieving an excellent recognition rate of 98.843% (108 classes, 3 training images and 4 test images for each class, hence there are 324 training images and 432 test images with no overlap between the training and test images). On the same training and testing set Libor Masek's algorithm can reach a recognition rate of 97.917%.

Libor Masek, Peter Kovesi. MATLAB Source Code for a Biometric Identification System Based on Iris Patterns. The School of Computer Science and Software Engineering, The University of Western Australia, 2003.

Index Terms: Matlab, source, code, iris, recognition, dct, discrete cosine transform, Karhunen-Loeve Transform, KLT.

Release 1.0 Date 2008.11.18
Major features:

Iris Recognition . It Luigi Rosa mobile +39 3207214179