Iris image quality on enhanced biometric recognition performance

Abstract

Nowadays, recognition patterns play an important role in several applications, in which the iris recognition is widely developed in authentication systems today. For such systems models, it is necessary to use as input high quality images, which will reduce possible recognition errors. Thus, this article develops experimental tests to study the image quality on the performance of the iris recognition, using different quality metrics. Thus, experiments are conducted with different iris images and applying the Hamming distance algorithm as reference measurement to accept or denied an user authentication. To this end the OSIRIS platform was used in the tests, because it permits to calculate the Hamming Distance between two binary codes. Based on the results obtained in the tests using different metrics, can be inferred that the image quality has a considerable impact on the performance of an iris recognition system. Therefore, the image capture process is an important step.