Article

Article title CLUSTERING OF IRIS TEMPLATES FOR OPTIMIZING MATCHING AGAINST BIG DATABASES
Authors I.V. Simonenko, I.A. Matveev
Section SECTION IV. MATHEMATICAL METHODS OF AN ARTIFICIAL INTELLECT
Month, Year 06, 2012 @en
Index UDC 004.93’14, 51-76, 57.087.1
DOI
Abstract Biometric identification system workflow can be split into two stages: determining biometric features (i.e. template) and matching this template against previously registered ones contained in database. The matching against big databases can take considerable, unacceptable time. A way of reducing matching time is proposed by means of clustering some group of template samples. The clusters break the space of templates into parts and yield an order of fetching of templates for matching, which is better than default straightforward fetching. In the tests, applying the method to public domain iris images resulted in reduction of number of comparisons in 1.5-2.5 times.

Download PDF

Keywords Iris identification; clusterization.
References 1. Pato J.N., Millett L.I. Biometric recognition: challenges and opportunities // National Research Council: Whither Biometrics Committee. – 2010.
2. Future Challenges based on the Multiple Biometric Grand Challenge // NIST Information Access Division: Multiple Biometric Grand Challenge. – February 2010. – Web: http://face.nist.gov.
3. Hao F., Daugman J., Zielinski P. A fast search algorithm for a large fuzzy database // IEEE Trans. Information Forensics and Security. – № 3 (2). – P. 203-212.
4. Daugman J. How iris recognition works // IEEE Trans. CSVT 14(1). – P. 21-30.
5. Yu L., Wang K., Zhang D. Coarse Iris Classification Based on Box-Counting Method // Proc. IEEE ICIP 2005. 11-14 September. – Vol. 3. – P. 301-304.
6. Vatsa M., Singh R., Noore A. Improving Iris Recognition Performance Using Segmentation, Quality Enhancement, Match Score Fusion, and Indexing // IEEE Trans. SMC. B-Cybern. – 2008. – № 38 (4). – P. 1021-1035.
7. Ким Дж.-О., Мьюллер Ч.У., Клекка У.Р., Олдендерфер М.С., Блэшфилд Р.К. Факторный, дискриминантный и кластерный анализ / Под ред. И.С. Енюкова. – М.:Финансы и статистика, 1989. – 215 с.
8. Monro D. University of Bath Iris Image Database. 2005. http://www.bath.ac.uk/eleceng/ research/sipg/irisweb/.
9. Chinese academy of sciences institute of automation (CASIA), CASIA Iris image database.2005. http://www.cbsr.ia.ac.cn/IrisDatabase.htm.

Comments are closed.