Neurotechnology’s iris recognition algorithms were among the top scorers in the NIST Iris Exchange (IREX) IX evaluation.
The Iris Exchange (IREX) IX is an evaluation of automated iris recognition algorithms.
The Iris Exchange (IREX) was initiated at NIST to expand iris recognition capabilities and support a marketplace of iris-based applications based on standardized interoperable iris imagery.
The work was initially conducted to support of the ISO/IEC 19794-6 standard and later the ANSI/NIST ITL 1-2007 Type 17 standard.
Subsequently IREX has included dedicated activities in support of:
- Iris image quality assessment and standardization (IREX II)
- Performance of one-to-many algorithms (IREX III + IV)
- Compression (IREX IV)
- Instructions to personnel involved in iris image collection (IREX V), and
- Time dependence of iris recognition (IREX VI)
The first part of the evaluation is a performance test of both verification (one-to-one) and identification (one-to-many) recognition algorithms over operational test data.
Thirteen commercial and research institutions developers submitted iris recognition algorithms for testing for this a large-scale identification task, which is more than any previous IREX evaluation.
Performance was measured for 46 matching algorithms over a set of approximately 700K field-collected iris images.
Neurotechnology’s algorithm had the second best accuracy among all tested iris recognition algorithms for both verification (1:1) and identification (1:N) scenarios.
Additionally, their accelerated iris identification algorithm was nearly 50 times faster than any other matcher in the IREX IX evaluation according to NIST’s report.
“Once again the NIST IREX evaluation proved that our ongoing improvements in both accuracy and speed provide our customers with the best-performance solutions for real-time iris recognition biometric systems at any scale,” said Dr. Justas Kranauskas, head of the biometric research department for Neurotechnology.
The only vendor to achieve higher accuracy results in this latest IREX evaluation processes iris images seven times slower, creates a roughly two-and-a-half times larger template and matches irises two times slower than Neurotechnology’s VeriEye and MegaMatcher iris recognition algorithm in a verification scenario.
Neurotechnology’s MegaMatcher product line includes:
- MegaMatcher SDK, a multibiometric SDK for large scale systems;
- MegaMatcher Accelerator, the world’s fastest biometric matching engine, and the MegaMatcher Automated Biometric Identification System (ABIS), a turnkey solution
The MegaMatcher line covers the following biometric modalities as well as SDKs for them: fingerprint (VeriFinger), face (VeriLook), eye iris (VeriEye) and voice (VeriSpeak) modalities.
MegaMatcher is designed for large-scale AFIS and multi-biometric systems developers. The technology ensures high reliability and speed of biometric identification even when using large databases.
Available as a software development kit that allows development of large-scale single- or multi-biometric fingerprint, iris, face, voice or palm print identification products for Microsoft Windows, Linux, Mac OS X, iOS and Android platforms.
MegaMatcher SDK includes all of the following modalities:
- MegaMatcher Accelerator provides high speed, high volume biometric identification suitable for national-scale projects.
- It is available in several editions, including the recently introduced Extreme edition.
- The Extreme edition is capable of matching 1.2 billion fingerprints per second, 1.2 billion faces per second or 700 million irises per second on a single server with a GPU which make it fastest biometrical matcher in the market.
While the MegaMatcher SDK is intended for system integrators, MegaMatcher ABIS is Neurotechnology’s turnkey multi-biometric solution for large-scale identification projects.
It is a modular and customizable solution that provides services for high- performance multi-biometric systems with multiple transactions using any combination of fingerprint, face or eye iris biometric modalities.
It is available as a complete system and ready for the deployment in national-scale projects.
MegaMatcher SDK Features and Capabilities:
- Proven in national-scale projects, including passport issuance and voter deduplication.
- NIST MINEX-compliant fingerprint engine, NIST IREX proven iris engine.
- High performance matching for national-scale systems with MegaMatcher Accelerator.
- Fingerprints, irises and faces can be matched on smart cards using MegaMatcher On Card.
- Includes fingerprint, iris, face, voice and palm print modalities.
- Rolled, flat and latent fingerprint matching.
- BioAPI 2.0 and other ANSI and ISO biometric standards support.
- ICAO requirements compliancy check for face images.
- Effective price/performance ratio, flexible licensing and free customer support.
In the NIST IREX identification scenario, the MegaMatcher Accelerator iris matching algorithm from Neurotechnology was 50 to 1000 times faster than any other algorithm in the top accuracy range.
MegaMatcher Accelerator is a hardware/software solution that provides high speed, high volume biometric identification for national-scale projects.
The Extreme and Extended versions are designed to run on server hardware and perform fast biometric template matching on the server-side of a large-scale AFIS or multi-modal system.
The Standard version and Development Edition are intended to be run on a common PC.
MegaMatcher Accelerator software licenses are available for new and existing MegaMatcher Extended SDK customers.
MegaMatcher Accelerator Features and Capabilities:
- Proven in national-scale projects, including voter registration and national ID.
- Available as a ready-to-use biometric solution with server hardware
or as biometric software that will run on server hardware or a common PC.
- Up to 1,200,000,000 fingerprints or 1,200,000,000 faces or 700,000,000 irises per second matching speed on a single unit.
- Fingerprint, iris, face and voice modalities supported.
- Scalable cluster architecture.
- Support for ISO & ANSI fingerprint template standards.
- Suitable for duplicates search.
Neurotechnology also submitted top performing iris recognition algorithms in 2009, 2012 and 2013 IREX evaluations.
As in past evaluations, Neurotechnology’s algorithms continue to be among the most accurate for iris recognition while also being the fastest by a wide margin.
Also Neurotechnology has consistently achieved industry-leading results from NIST for fingerprint algorithms.
In the latest MINEX III evaluation, Neurotechnology’s fingerprint template generator algorithm ranked first in the interoperability category and the fingerprint matching algorithm ranked second.
When combined, the two offer the most accurate, high speed fingerprint recognition system.
VeriFinger is a fingerprint identification technology designed for biometric systems developers and integrators.
The technology assures system performance with fast, reliable fingerprint matching in 1-to-1 and 1-to-many modes.
Available as a software development kit that allows development of stand-alone and Web-based solutions on Microsoft Windows, Linux, Mac OS X, iOS and Android platforms.
Neurotechnology offers a full line of products for iris (VeriEye), fingerprint (VeriFinger), face (VeriLook) and voice (VeriSpeak) modalities as well as MegaMatcher SDK multibiometric technology, which includes all modalities in any combination.
Neurotechnology is a recognized developer of high-precision algorithms and software based on deep neural network (DNN) and other AI-related technologies.
Drawing from years of academic research in the fields of neuroinformatics, image processing and pattern recognition, Neurotechnology has released more than 130 products and version upgrades.