Quality Metrics for Biometrics

The accuracy of a biometrics system increases with the quality of the sample used for identification. All current iris biometrics systems capture multiple samples that must be processed to identify a single, ideal image to be used for identification. Many metrics exist to evaluate the quality of an iris image. This thesis evaluates current metrics and introduces a method for determining the ideal iris image from a set of iris images by using the Mean Self-Match algorithm to examine the set of true matches. This proposed method is shown to outperform other methods currently used for selecting an ideal image from a set of iris images. The application of this method to face biometrics is also examined.