Home // ICCGI 2012, The Seventh International Multi-Conference on Computing in the Global Information Technology // View article

Does Collarette of Iris Work for Recognizing Persons?

Authors:
Ren-He Jeng
Wen-Shiung Chen
Lili Hsieh

Keywords: Biometrics; Iris Recognition; Iris Localization; Iris Normalization; Collarette

Abstract:
In recent years, iris has been extensively discussed in the field of biometrics. An iris recognition system has three main stages such as image preprocessing, feature extraction and template matching. Since eyelid and eyelashes act as a kind of armour that protect the eye from harm, it makes iris localization inaccurate in image pre-processing step. A novel method is proposed to locate iris radius based on collarette of iris muscle. The collarette is the thickest region of the iris, separating the pupillary portion from the ciliary portion. Some researches suggest that the parameters of iris normalization algorithm adopts collarette to replace iris outer radius. However, reducing the normalization radius of iris will deform normalized iris image and result in lose of iris feature information. In this paper, we present our experiments by adopting different iris radii and different normalization algorithms in iris recognition system. We also propose an iris localization method and a collarette localization method. In feature extraction, we adopt the Gabor wavelet filter to extract local texture features from iris images. All experiments are tested on UBIRIS Sessao.1 and CASIA.v1 databases. The experimental results show that the proposed approach has achieved a high accuracy of 96%.

Pages: 89 to 94

Copyright: Copyright (c) IARIA, 2012

Publication date: June 24, 2012

Published in: conference

ISSN: 2308-4529

ISBN: 978-1-61208-202-8

Location: Venice, Italy

Dates: from June 24, 2012 to June 29, 2012