A New Method for Eye Printing

The aim of this lecture is to present a new method for eye and more precisely iris printing and recognition. Iris printing is used for identification purposes. The currently used methods for iris printing have major constraints which limit their applications. We present a method which alleviates these constraints and therefore may have significant new applications. The method first separates the iris from the rest of the image. For analysis of the iris a multidimensional feature vector is introduced. The feature vector is constructed using the geometry of surfaces obtained from higher order statistics and the frequency domain analysis of the statistics. The method has been applied to a number of test cases and its performance has been comparable to the best known methods without being encumbered by unnecessary constraints.