Proceedings of the IEEK Conference (대한전자공학회:학술대회논문집)
- 2008.06a
- /
- Pages.863-864
- /
- 2008
Analysis on Iris Image Degradation Factors
홍채 인식 성능에 영향을 미치는 화질 저하 요인 분석
- Yoon, So-Weon (Biometrics Engineering Research Center School of Electrical and Electronic Engineering Yonsei University) ;
- Kim, Jai-Hie (Biometrics Engineering Research Center School of Electrical and Electronic Engineering Yonsei University)
- Published : 2008.06.18
Abstract
To predict the iris matching performance and guarantee its reliability, image quality measure prior to matching is desired. An analysis on iris image degradation factors which deteriorate matching performance is a basic step for iris image quality measure. We considered five degradation factors-white-out, black-out, noise, blur, and occlusion by specular reflection-which happen generally during the iris image acquisition process. Experimental results show that noise and white-out degraded the EER most significantly, while others on EER were either insignificant or degradation images resulted in even better performance in some cases of blur. This means that degradation factors that affect the performance can be different from those based on human perception or image degradation evaluation.
Keywords