Non-circular iris contours localization in the visible wavelength eye images. (August 2017)

• Record Type: Journal Article
• Authors: Jan, Farmanullah
• Is Part Of: Computers & electrical engineering. Volume 62(2017)
• Page Start: 166
• Page End: 177
• Publication Date: 2017-08
• Subjects: Biometric systems -- Machine vision -- Iris recognition -- Iris localization -- Pattern recognition; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2016.11.031
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Effective non-circular iris contours localization algorithm. Validated on the visible wavelength iris databases. Validated on the near infra-red iris databases. It is real-time applicable. Abstract: Contemporary iris localization schemes perform well for ideal eye images generally acquired using near infrared (NIR) illumination, but perform poorly for non-ideal eye images acquired using Visible Wavelength (VW) illumination. Non-ideal eye images contain noise such as non-uniform illumination, defocus, etc. This study proposes a reliable iris localization algorithm, in particular for VW eye images. First, it suppresses light reflections and sharp intensity variations in the Red-component of input color eye image. Next, it marks a set of potential pixels (ROIpixels) suitable as the iris center's candidates. Using ROIpixels, it localizes potential iris location using a hybrid of Integro-differential operator (IDO) and image intensity. Then, it reuses IDO and iris/pupil's geometry to localize pupil within iris. Finally, it uses the Fourier series based scheme to regularize iris contours. Experimental results obtained on UBIRIS V1.0, MMU V1.0, and IITD V1.0 iris databases demonstrate its superior performance compared with contemporary schemes. Graphical abstract: