Publications:Iris Boundaries Segmentation Using the Generalized Structure Tensor : A Study on the Effects of Image Degradation

From ISLAB/CAISR

Do not edit this section

"CAISR research program of the Swedish Knowledge Foundation;EU BBfor2 Marie Curie Initial Training Network "Bayesian Biometrics for Forensics" (FP7-ITN-238803);EU COST Action IC1106 "Integrating Biometrics and Forensics for the Digital Age";EU FP7 Marie Curie Intra-European Fellowship "FP7-PEOPLE-2009-IEF-254261-BIO-DISTANCE";Swedish Research Council Postdoctoral Grant "2009-7215"" cannot be used as a page name in this wiki.

Keep all hand-made modifications below

Title Iris Boundaries Segmentation Using the Generalized Structure Tensor : A Study on the Effects of Image Degradation
Author Fernando Alonso-Fernandez and Josef Bigun
Year 2012
PublicationType Conference Paper
Journal
HostPublication Biometrics: Theory, Applications and Systems (BTAS), 2012 IEEE Fifth International Conference on
Conference The IEEE Fifth International Conference on Biometrics: Theory, Applications and Systems (BTAS 2012), Washington DC, September 23-26, 2012
DOI http://dx.doi.org/10.1109/BTAS.2012.6374610
Diva url http://hh.diva-portal.org/smash/record.jsf?searchId=1&pid=diva2:545745
Abstract We present a new iris segmentation algorithm based onthe Generalized Structure Tensor (GST), which also includesan eyelid detection step. It is compared with traditionalsegmentation systems based on Hough transformand integro-differential operators. Results are given usingthe CASIA-IrisV3-Interval database. Segmentation performanceunder different degrees of image defocus and motionblur is also evaluated. Reported results shows the effectivenessof the proposed algorithm, with similar performancethan the others in pupil detection, and clearly betterperformance for sclera detection for all levels of degradation.Verification results using 1D Log-Gabor wavelets arealso given, showing the benefits of the eyelids removal step.These results point out the validity of the GST as an alternativeto other iris segmentation systems. © 2012 IEEE.