Advertisement
No two irises are exactly alike, and although they may appear to be round,no iris is perfectly circular. Unlike some iris identification systems thatapproximate the iris shape as a circle, Neurotechnology's new irissegmentation algorithm can detect the boundaries of irises that are notperfectly circular, are partly occluded by eyelids or eyelashes or deformed inshape because the eye is looking away from the image capture device. Thisenables VeriEye to more precisely extract all visible iris texture andsignificantly enhance iris recognition accuracy and reliability. Iris textureis modeled through unique features that remain true even with varying contrastand illumination levels. VeriEye takes less than a half-second to create arecord of unique iris features from an unsegmented human eye image.
Advertisement
The new iris matching algorithm in VeriEye rapidly compares two sets ofiris features and is similarly capable of detecting and compensating forobstructions on iris texture and in deformed "looking away" iris images.Matching speeds against existing iris records can range from 50,000 to 150,000irises per second, depending on configuration.
"We are very pleased to add VeriEye SDK for iris recognition to our lineof biometric fingerprint, face and object recognition technologies," saidJustas Kranauskas, Neurotechnology's iris identification project lead. "Havinga full suite of biometric recognition modalities allows us to provide a best-fit solution for each customer."
VeriEye provides a high degree of accuracy and reliability. When tested onthe publicly available dataset ICE2005 Exp 1 (NIST ICE-1), Neurotechnology'sVeriEye iris recognition algorithm achieved the following results at amatching speed of 50,000 iris templates per second:
These are the best known results ever to have been achieved on thisdatabase. Similar results are achieved on other datasets.
VeriEye Standard SDK and VeriEye Extended SDK are available now withhighly competitive licensing options through Neurotechnology or fromdistributors worldwide. A 30-day trial version with full functionality is alsoavailable for download. For more information and downloads, go to:http://www.neurotechnology.com.
About Neurotechnology
Neurotechnology is a provider of high-precision biometric fingerprint,face and iris identification algorithms, object recognition technology andsoftware development products. More than 1900 system integrators, securitycompanies and hardware providers integrate Neurotechnology's algorithms andsoftware development technologies into their own products, with millions ofcustomer installations worldwide.
Neurotechnology's identification algorithms have consistently earned thehighest honors in some of the industry's most rigorous competitions, includingthe National Institute of Standards & Technology (NIST)'s Fingerprint VendorTechnology Evaluation (FpVTE) and the International Fingerprint VerificationCompetitions (FVC).
Drawing from years of academic research in the fields of neuroinformatics,image processing and pattern recognition, Neurotechnology was founded in 1990in Vilnius, Lithuania under the name Neurotechnologija and released its firstfingerprint identification system in 1991. Since that t
