Software combines finger and face identification.
January 31, 2007 -
Featuring fused matching algorithm, MegaMatcher 2.0 SDK is suited for development of large-scale automated fingerprint identification systems and multi-biometric face-fingerprint identification systems. Algorithm allows technologies to work together to provide 1:N matching and can produce up to 400,000 matches/sec on single processor PC. Fingerprint algorithm is tolerant to fingerprint translation, rotation, and deformation and has ability to match between rolled and flat fingerprints.
(Archive News Story - Products mentioned in this Archive News Story may or may not be available from the manufacturer.)
|Original Press release |
Ateities str. 10
Vilnius, , LT-08303
MegaMatcher 2.0 SDK is Now Available for the Development of Large-Scale AFIS and Multi-Biometric Face-Fingerprint Identification Systems
New Fused Matching Algorithm Can Produce Up to 400,000 Matches Per Second With High Reliability On a Single PC
Vilnius, Lithuania - January 29, 2007 - Neurotechnologija, a provider of high-precision biometric identification technologies, today announced the availability of MegaMatcher 2.0 SDK for the development of large-scale automated fingerprint identification systems (AFIS) and multi-biometric face-fingerprint identification systems. MegaMatcher 2.0 includes both fingerprint and face identification engines with a fusion algorithm that allows the two technologies to work together to provide very fast 1:N (1 to many) matching with even higher reliability than AFIS or facial recognition alone. MegaMatcher's powerful fused algorithm can produce up to 400,000 matches per second on a single processor PC; and with MegaMatcher 2.0's fault-tolerant, scalable cluster software, this number can be multiplied across multiple PCs to perform extremely fast, parallel fingerprint and face matching using databases of practically unlimited size.
With the increasing demand for automated biometrical identification systems for both forensic and civil applications, such as crime scene investigations, border crossings, passport and visa documentation, election control systems and even credit card transactions, the need for both fast and accurate identification is ever more critical. While AFIS have been widely used for the past two decades, multi-biometric systems that incorporate both face and fingerprint identification offer a number of advantages. Using the multi-biometric technology in MegaMatcher 2.0, developers and integrators can create systems where both face and fingerprint can be scanned at the same time using inexpensive hardware such as a fingerprint scanner and a simple Webcam or photo scanner (for example, scanning a passport photo). A multi-biometric system based on MegaMatcher 2.0 provides both greater convenience and matching speed for the user and a much higher level of security, even when using very large databases, because the fusion of both face and fingerprint provides unambiguous identification.
The face and fingerprint algorithms in MegaMatcher 2.0 are Neurotechnologija's most powerful algorithms to date and include a number of enhanced features to ensure higher quality and accuracy in both enrollment and matching. New fingerprint measurement tools improve recognition by enabling poor fingerprints to be rejected at the enrollment stage. The new template generalization tool facilitates faster, more accurate matching by enabling a much higher quality template to be created from just a few fingerprints or faces. The MegaMatcher 2.0 fingerprint algorithm is tolerant to fingerprint translation, rotation and deformation and has the ability to match between rolled and flat fingerprints.
Compared to the fingerprint-only algorithm in MegaMatcher 1.1, the MegaMatcher 2.0 algorithm offers considerably higher reliability in both fingerprint and multi-biometric modes. When using fingerprints only, the False Rejection Rate (FRR) in MegaMatcher 2.0 is up to six times lower than MegaMatcher 1.1 when set to the same False Acceptance Rate (FAR) level. In the multi-biometric mode, at the same FAR level, the FRR decreases by 10 to 100 times compared to the previous version.
The fast MegaMatcher 2.0 algorithm also provides significant economies of scale (and hardware price) when using the system in a cluster. For example, an application that required 100 PC cluster nodes for MegaMatcher 1.1 would require only 15 nodes when using MegaMatcher 2.0.
Enhancements to the MegaMatcher 2.0 SDK make it easier to work with fingerprints both during and after enrollment. When multiple fingerprints have been scanned at the same time, the new fingerprint segmentation functionality in MegaMatcher 2.0 can automatically segment them into separate fingerprints. And for forensic applications, MegaMatcher 2.0 includes an API for the editing of latent fingerprint templates, such as those collected from crime scenes.
"The potential uses for multi-biometric capabilities with a large-scale AFIS are enormous," said Algimantas Malickas, CEO of Neurotechnologija. "So with the new MegaMatcher 2.0 SDK we've worked hard to make it easier than ever for developers and integrators to put this powerful functionality in their systems."
MegaMatcher 2.0 is fully compatible with other Neurotechnologija products, including VeriFinger, VeriLook and the recently announced embedded development kits FaceCell and FingerCell 2.0. All components of MegaMatcher 2.0 can be used on both Microsoft Windows and Linux operating systems, and the SDK also includes an improved component set for simpler and faster integration.
MegaMatcher 2.0 includes .NET components for rapid development of client side software. And to ensure compatibility with other companies' AFIS software, a WSQ library and modules for conversion between the MegaMatcher template and ANSI standards (ANSI/NIST ITL-1-2000 and ANSI/INCITS 378-2004) are also included. MegaMatcher 2.0 supports CrossMatch, Digital Persona and more than 20 other fingerprint scanners.
MegaMatcher 2.0 is available now with highly competitive licensing options through Neurotechnologija or from distributors worldwide. A 30-day trial version with full functionality is also available for download. For more information and downloads, go to: http://www.neurotechnologija.com.
Neurotechnologija provides biometric fingerprint and face identification algorithms and software development products to security companies, system integrators and hardware manufacturers. More than 1000 system integrators and fingerprint sensor providers worldwide integrate Neurotechnologija's technologies into their own products.
Drawing from years of academic research in the fields of neuroinformatics, image processing and pattern recognition, Neurotechnologija was founded in 1990 in Vilnius, Lithuania and released its first fingerprint identification system in 1991. Since that time Neurotechnologija has released more than 30 products and version upgrades for identification and verification of personal identity. For more information, visit: http://www.neurotechnologija.com.