Iris Segmentation for Recognition using Local Statistics
Robert W. Ives, Lauren R. Kennell, R.M. Gaunt, Delores M. Etter · 2006
Iris recognition is one of the more commonly used biometrics for recognition due to its accuracy. One of the first steps in iris recognition is to segment the iris from the rest of the image for further processing. This paper investigates the use of local statistics to find the pupillary (inner) and limbic (outer) boundaries of the iris. In particular, the local standard deviation and local kurtosis have shown to be useful in this respect. The methodology and results using the University of Bath and the United States Naval Academy iris databases are presented.