A new probabilistic Iris Quality Measure for comprehensive noise detection

In this article, we present a novel probabilistic iris quality measure based on a Gaussian mixture model (GMM). We compare its behavior to that of other standard iris quality metrics on two different types of noise which can corrupt the iris texture: occlusions and blurring. In the case of occlusions, we compare our GMM-based quality measure to an active contour method for eyelids and eyelashes detection. Finally, in the case of iris blurring, we compare our quality measure to a standard method based on Fourier transform and wavelets. For the latter, we have developed a new method to detect blur suitable for iris images. In our tests, we have used the ICE 2005 database and OSIRIS, an iris reference system based on the classical approach proposed by Daugman and developed in the framework of BioSecure European Network of Excellence for comparative evaluation purposes. Experiments show a significant improvement of performance when our GMM- based quality measure is used instead of the classical methods above mentioned. In particular, results show that this probabilistic quality measure based on a GMM trained on good quality images is independent from the kind of noise involved.