Due to its distinctiveness, the human iris is a popular biometric feature used to identity a person with high accuracy. The Grand Challenge in iris recognition is to have an effective algorithm for subject verification or identification under a broad range of image and environmental conditions. This paper presents both baseline performance results and an enhanced version of VASIR (Video-based Automated System for Iris Recognition) as a response the challenge. We describe the details of the VASIR procedure and demonstrate its superiority over the IrisBEE algorithm. Finally, the image quality scores and how they relate to VASIRs performance are examined.
Citation: NIST Interagency/Internal Report (NISTIR) - 7777Report Number:
NIST Pub Series: NIST Interagency/Internal Report (NISTIR)
Pub Type: NIST Pubs
biometrics, iris recognition, VASIR, baseline, benchmark, Hamming distance, image processing