Neurotechnology’s fingerprint algorithm maintains top Nist ranking

Neurotechnology’s fingerprint algorithm maintains top Nist ranking

Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies, today announced that the company has confirmed the first place position in the NIST Proprietary Fingerprint Template (PFT) III evaluation results released on March 7, 2022. Since 2019, Neurotechnology’s fingerprint recognition algorithms have held the number one position in NIST PFT evaluations as the company has continued to innovate and improve both the speed and accuracy of their biometric algorithms.

The highly accredited National Institute of Standards and Technology (NIST) has, with its Proprietary Fingerprint Template (PFT) evaluation, provided the largest and most recognized ongoing assessment of fingerprint verification available today. Since fingerprint templates are not necessarily standardized across all vendors, each algorithm developer uses their latest technology to showcase the full potential of their indvidual algorithms. Neurotechnology’s top position in PFT III demonstrates the fully enhanced capabilities of the company’s latest R&D advancements.

The Proprietary Fingerprint Template evaluation determines the algorithmic capabilities of one-to-one fingerprint verification and assesses the accuracy of final stage fingerprint matchers used in one-to-many Automated Fingerprint Identification System (AFIS) searches. The most recent generations of the PFT evaluation cover:

PFT II: one-to-one fingerprint verification evaluation. Active between 2010 and 2019, PFT II was used to evaluate plain versus plain, plain versus rolled, and rolled versus rolled fingerprint verification scenarios and reported template extraction times, template sizes, and match times.
PFT III: the continuation of the previous PFT generation evaluations now includes several PFT III-specific datasets.
To provide a more comprehensive comparison of algorithmic function, NIST included in this current evaluation the accuracies that were being tested in the PFT 2003 and PFT II evaluations. This process provided a comparison of the algorithm submissions across all current and former PFT evaluations.

Neurotechnology strengthens NIST FRVT positions

Neurotechnology strengthens NIST FRVT positions

Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies, today announced that the company’s latest face recognition algorithm showed significant improvements among the top NIST FRVT testing results released on January 13, 2022.

The Face Recognition Vendor Test (FRVT) conducted by the National Institute of Standards and Technology (NIST) is the most reliable series of large-scale, independent evaluations for face recognition algorithms in verification (1:1) and identification (1:N) scenarios. Immense datasets containing photos of faces are used during the evaluation to measure the performance of face recognition algorithms developed worldwide.

The new face recognition algorithm from Neurotechnology has demonstrated significant advancement in both FRVT 1:1 and FRVT 1:N NIST testing, showing comprehensive performance across identification and verification testing scenarios.

“Consistency and dedication are crucial to our sustained R&D accomplishments,” said Evaldas Borcovas, biometric research lead at Neurotechnology. “Previously our team achieved the best algorithm accuracy in fingerprint recognition evaluations, and now we are seeking to do the same in face recognition evaluations. Based on our experience, and these latest algorithm results, I am confident that we are moving in the right direction.”

Neurotechnology - FRVT 1x1 PFT.PNG

In the FRVT 1:1 Verification evaluations, Neurotechnology’s face recognition algorithm showed significant performance improvements, including:

  • In the top 3% of most accurate results for border control supervised (Visa Border, Border) and unsupervised (Kiosk) scenarios among 702 submissions by 255 providers.
  • Among the top 3% of algorithms for accuracy with masks from a total of 319 entries.

Nuerotechnology - FRVT 1xN PFT.PNG

The face recognition algorithm also showed significant performance improvements in the FRVT 1:N Identification evaluations, including:

  • In the top 4% of the leading results matching frontal and profile mugshots scenarios among 341 submissions by 93 different providers.
  • Top results among border control supervised (Visa vs Border, Border vs Border ΔT ≥ 10 YRS) and unsupervised (Visa vs Kiosk) scenarios.
  • Leading score by template size. Considering the template size, the algorithm showed the best results among all other submissions with the same template size.

These results in the NIST FRVT demonstrate that the latest face recognition algorithm from Neurotechnology continues the company’s strong track record of providing face recognition products that are among the top performing solutions for some of the most common situations in civilian and law enforcement scenarios, as well as offering industry-leading efficiency by template size, extraction, and matching speed performance.


Neurotechnology launches latest MegaMatcher 12.3

Neurotechnology launches latest MegaMatcher 12.3

Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies, today announced the release of the new MegaMatcher 12.3 multi-biometric product line, including updates to the MegaMatcher software development kit (SDK) and to MegaMatcher Accelerator – a combined software and hardware solution that provides high-speed, high-volume biometric identification for national-scale projects. The latest versions include enhanced facial and iris recognition algorithms with improved liveness detection, a new voice recognition algorithm and a new inference engine that provides significantly better speed and performance across all biometric modalities.

“Every day our team aims to innovate technologies that make our products more accurate and robust while also being faster and less complex for our customers to use,” said Evaldas Borcovas, biometric research lead at Neurotechnology. “This latest version of MegaMatcher exemplifies these efforts, providing greater accuracy in verification and identification processes while enabling our customers’ systems to work faster.”

The MegaMatcher product line includes Neurotechnology’s top-ranked biometric algorithms providing high recognition and identification accuracy across fingerprint, face, iris, palm print and voice biometric modalities that can be used individually or in any combination.

The latest enhancements to the MegaMatcher 12.3 product line include:

A new Intel Inference Engine provides better all-around performance, particularly for extraction operations. Support for Mac M1 (ARM) with neural network framework is now significantly faster than any previous versions running on the macOS.

Fingerprint. The new fingerprint algorithm includes support for the latest NIST Fingerprint Image Quality (NFIQ) 2.1 biometric standard, offering a higher degree of compatibility and flexible application.
Passive Face Liveness Detection. Newly introduced passive face liveness algorithm (also known as Presentation Attack Detection – PAD) establishes a higher degree of fraud prevention in mobile and dynamic situations.
Face. Additional updates to the face algorithm include an improved face extraction algorithm, better face detection and significantly improved detection of specific facial attributes including: gender, beard, mustache, hat, blink, mouth open, smile, glasses and dark glasses. A new attribute for glasses with a heavy frame is also introduced.
Iris Liveness Detection. Upgraded iris liveness algorithm (PAD) brings new improvements to eye activity detection and potential deceit risks assessment.

Voice. An entirely new algorithm introduces Neurotechnology’s advanced capabilities for voice recognition with multiple times better EER results.

MegaMatcher Accelerator. In addition to the proprietary API, version 12.3 also adds support for gRPC API (HTTP 2.0), making it easier to add security, load balancers and enabling the product to be used with any language (including Python and PHP) without native C components.

Face Verification 12 from Neurotechnology extends facial authentication capabilities

Face Verification 12 from Neurotechnology extends facial authentication capabilities

Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies, today announced the release of the new Face Verification system. Face Verification 12 is designed for the integration of facial authentication and liveness detection into PC, mobile and web applications for digital onboarding, payment, banking, telecommunications and other face recognition uses on personal devices.

The new facial recognition algorithm in Face Verification 12 features the same simple APIs as the previous version for enrollment, liveness checks, authentication and ISO 19794-5 quality checks while providing even better accuracy for both face recognition and liveness detection. It also includes a new encrypted biometric template format with a compact dimension. This compact template size makes it easier to store biometric data on small micro controllers or even include it in a QR code with other data, improving the overall versatility of the product for use in a wider range of environments.

The new version introduces a web service component that provides server-based performance for all operations that were previously available only for use on mobile devices and PCs. The web service is ready to be deployed on the server of a system integrator or of an end customer, and the operations can be performed through a WebRTC and REST API while processing live streaming from a camera in real time.

Face Verification 12 adds an improved passive mode to both mobile and server components. The new passive modality doesn’t require the user to perform any active action, and it is able to detect various types of spoofing attacks, including masks, videos, photos and others. The new liveness modality adds to the passive and active modalities in the previous Face Verification SDK, including an improved blinking detection algorithm.

The new SDK component of the Face Verification system also allows users to export the face biometric data that is generated and use it to perform enrollment and verification on smart cards that integrate the new MegaMatcher On Card 12 algorithm released by Neurotechnology on November 29, 2021. This smart card technology performs on-card face template comparisons and can be used as an additional authentication factor for applications that require a very high degree of security. The combination of Face Verification and MegaMatcher On Card forms a unique product ecosystem in the identity verification market and provides even greater flexibility and value.

“We see an increasing demand for face recognition technology for consumer applications,” said Antonello Mincone, business development manager for Neurotechnology. “However the requirements vary widely, depending on the location where data is stored, where the biometric checks are perfomed and the different privacy and security regulations involved. Face Verification 12 responds to this demand with a new web service component and smart card compatibility that expand the architectural capabilities for face authentication. It allows system integrators to easily implement and deploy different enrollment and authentication workflows for mobile, web and hybrid applications,” Mincone added.

Face Verification 12 has a convenient licensing model that accounts for only the number of persons that have effectively used the deployed system, regardless of whether the enrollment or liveness check took place on mobile or web. So the same biometric data, once enrolled, can later be verified in a variety of authentication scenarios and without extra operational costs.

MegaMatcher On Card 12 from Neurotechnology features new card algorithms

MegaMatcher On Card 12 from Neurotechnology features new card algorithms

 Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies, today announced the release of the new MegaMatcher On Card software development kit (SDK). MegaMatcher On Card is an SDK for developing applications that feature biometric comparisons directly within the microcontroller of smart cards, and it includes optimized versions of the top ranked fingerprint, face and iris biometric algorithms from Neurotechnology.

The latest version of the SDK includes a new implementation of the face recognition algorithm with a biometric data template size lower than 256 bytes. This small size helps to save storage space on microcontrollers and enables the transmission of the face template to smart cards in a single data unit. This results in faster authentication speed and makes it easy to add support for face biometrics as a modality in third party smart card applications that were previously developed only for fingerprints.

MegaMatcher On Card 12 introduces the 1-to-N extension of the fingerprint library for the automatic recognition of multiple fingerprints on the same smart card, making it possible to verify the card owner directly through the use of any of the fingerprint data stored in the card. This is more convenient for end users and for system integrators who don’t have to implement complex mechanisms to require a specific fingerprint for verification.

“There are already more than 130 million smart cards and secure elements worldwide that include our MegaMatcher On Card technology,” said Antonello Mincone, business development manager for Neurotechnology. “We are constantly improving our on-card technology and adding value for its different uses, as we do with all our algorithms and related products across multiple platforms,” Mincone added.

MegaMatcher On Card 12 also includes support, through code samples and tutorials, for NXP® JCOP 4 smart cards with a fingerprint 1-to-1 library from Neurotechnology.

Neurotechnology biometric voter system used in Ghana vote

Neurotechnology biometric voter system used in Ghana vote

Neurotechnology, a provider of deep-learning-based solutions and high-precision biometric identification and object recognition technologies, today announced that the company’s services and products, including the MegaMatcher Automatic Biometric Identification System (ABIS), were used for Ghana’s Biometric Voter Management System, providing voter registration, deduplication, adjudication, final voter list generation and verification during voting for the country’s general elections held on December 7, 2020.

Neurotechnology delivered the Biometric Voter Management System (Software) and provided related services for the Ghana Electoral Commission in record time – just nine months from start to finish – covering everything from initial voter registration to final voter list preparation and onsite support for the General Election.

Neurotechnology’s MegaMatcher ABIS provided deduplication for a total of 17,027,641 registrants who were eligible to vote in the general election. By matching fingerprints and/or facial biometrics for each registrant name, the system successfully identified 15,860 multiple registrations conducted by 7,890 unique individuals who attempted to register more than once using different names.

“We are highly satisfied with Neurotechnology’s software and services, which were provided in a short timeframe –nine months to be precise, from inception to completion,” stated Mrs. Jean Mensa, Chairperson for the Electoral Commission of Ghana. “The staff of Neurotech were highly professional and helpful.  They were responsive and accessible in that they were physically present on site throughout the duration of the exercise.

By providing highly accurate identification and verification of each registered voter, and by ensuring that each voter was only able to cast one vote, the MegaMatcher ABIS Biometric Voter Management System helped to ensure fair, transparent, credible and successful general elections in Ghana.

“We were pleased to work with the Electoral Commission of Ghana for their Biometric Voter Management System,” said Irmantas Naujikas, Director of Neurotechnology.


Neurotechnology releases app for

Neurotechnology releases app for

Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies, has released a mobile app that enables users to easily use image recognition models from their phone. enables users to build, train and deploy image recognition models without the need of an understanding of AI or deep learning. The app serves as a useful companion to the online platform, which itself has received a recent update.

The app, which is available on both the Google Play Store and the Apple App Store, brings a new level of convenience for users, as models can now be used immediately on the photos taken by a phone camera. app users have a wide range of pre-trained image recognition models to select from, or they can train a custom model using the web interface and make it available on their phone immediately. Additionally, the app enables users to:

  • Use the image similarity search function to upload a photo from a phone and then find similar photos within a dataset
  • Make predictions using photos uploaded from phone gallery
  • Use the AI-assisted labeling feature to save predictions as image labels in the dataset

The latest changes to the online platform focused on delivering the most user-friendly experience to date, through a range of new features and changes to the online dashboard and website. These changes included:

  • The introduction of a new pay-as-you-go wallet system that enables users to pay for only what they use on the platform
  • Object detection models became available for all registered users to build, train and deploy, whereas previously they had only been available for users on premium subscription plans
  • To benefit users with a large image dataset, users are now able to retrain full classification model networks, as well as being able to choose to automatically stop classification model training if there is no improvement for a specified amount of time
  • When using through the REST API, users are able to upload and delete images from a project
  • To broaden the accessibility of, new C# REST API code samples and Swagger specifications were added to provide users with the tools needed to incorporate into their own projects
  • In-depth user guides and video tutorials have been added to the website, to help every user make the most of the range of image recognition tools available on the platform