Automated Biometric Identification System

Turnkey multi-biometric solution for national-scale identification projects

MegaMatcher Automated Biometric Identification System is a complete system for the deployment of large-scale multi-biometric projects. The modular and customizable solution provides services for high-performance, scalable systems with multiple parallel transactions.

The solution is intended for national-scale projects, like biometric voter registration with records deduplication, passport issuing, border control, as well as other civil or criminal AFIS/ABIS.

Available as on-premise solution and as cloud service.

View data sheet Watch a Demo
Features and Capabilities
Technology proven in national-scale projects, like passport issuance and voter deduplication.
Fingerprint, face, iris and palmprint modalities supported.
Enrollment of biometric and biographic data with quality assessment.
Identity records lifecycle management with adjudication.
Galleries support for providing different agencies with specific access to the system.
Fast matching in verification, identification and deduplication modes.
Top accuracy proven at NIST MINEX III, PFT III, IREX 10 and FVC-onGoing.
Forensic analysis module for law enforcement applications available.
Scalable, modular architecture with high availability and fault tolerance
Cloud service available for making the system accessible from multiple platforms and locations.
Customization is possible for specific project needs.
MegaMatcher Automated Biometric Identification System (ABIS) is available as on-premise solution and as cloud service:
MegaMatcher ABIS 12.1 on-premise solution is designed as a complete system with all necessary components for deploying a large-scale biometric system. The solution includes ready-to-use services and applications for running on regular hardware.
MegaMatcher ABIS 12.1 cloud service is designed to make the system more accessible from multiple platforms and locations. The secure integrated cloud service provides functionality for enhancing its use as a solution for large and national-scale projects.
MegaMatcher ABIS 12.1 on-premise solution provides these capabilities for large-scale biometric identification systems:
Fingerprint, face, iris and palmprint modalities support. The system can accept fingerprint, face, iris and palmprint images for creating biometric templates and searching against the database. Each template can contain multiple fingerprints, irises, faces and/or palmprints. Proprietary fused matching algorithm provides high matching accuracy.
Biometric and biographic data enrollment. Person's biometric is captured in real time from stationary or mobile enrollment stations using the supported fingerprint/palmprint readers (150+ models supported) or iris cameras. Off-the-shelf cameras are used for ICAO compliant facial image capture. Quality assessment can be used for evaluation of the biometric data quality based on standard and proprietary algorithms. Document scanning is also available.
Identity records management. The whole lifecycle of identity records is managed, including registration, update and delete procedures. Complex and questionable biometric operations are resolved using the adjudication procedure, which can use automatic rulesets or forward the cases to human operators, who will review them using a specialized visual tool.
Galleries support. Identities can be optionally organized into separate and isolated galleries for grouping them by certain attributes (i.e. geographic region) and providing access to each of them for specific users. This way a single system can be used by multiple agencies for their specific needs.
Fast matching. The biometric matching can be performed in verification (1-to-1), identification (1-to-all) and deduplication modes (all-to-all). The fast and accurate matching is assured by proprietary algorithms which are able to perform up to 1.2 billion comparisons per second on each node of the system. The matching process can be tuned to detect anomalies. Multiple gallery support is optionally available for performing matching on isolated subsets of the biometric database.
Accuracy. MegaMatcher ABIS is based on award-winning biometric technologies. In 2019 NIST has judged that Neurotechnology's fingerprint algorithms are the most accurate high speed fingerprint recognition systems among all MINEX III participants. In the same year MegaMatcher palm print recognition algorithm has been recognized as the most accurate overall and fastest among the most accurate contenders at the FVC-onGoing evaluation. In 2020 and 2021 Neurotechnology's fingerprint algorithm earned first place at the FVC-onGoing and PFT III evaluations correspondingly. Also, in 2020 the iris recognition algorithm has been proven by NIST as the second most accurate at the IREX 10 evaluation.
Forensic analysis module. The module is designed for use in law enforcement applications and provides specific functionality including latent fingerprints processing.
Customization for project needs. Most components of MegaMatcher ABIS can be customized for specific project needs and restrictions. The customization may range from user interface translation into required language to modifying the system architecture.
Technology Awards

MegaMatcher Automated Biometric Identification System is based on awards-winning fingerprint and iris recognition technologies.

MINEX Evaluations by NIST

MINEX III evaluation was successfully passed in 2015. In 2019 Neurotechnology's fingerprint template generator algorithm has been ranked the first in the NIST MINEX interoperability category; the fingerprint matching algorithm has also been ranked as the front-runner in terms of interoperability and, when combined, the two have become the supreme accuracy, high speed fingerprint recognition system.
MINEX Ongoing evaluation was successfully passed in 2014. The second place in the Ongoing MINEX ranking for fingerprint matching algorithms was achieved. MegaMatcher technology was recognized by the NIST as fully MINEX compliant.

FVC-OnGoing Results

In 2020 MegaMatcher fingerprint recognition algorithm has shown the top result at the FVC-onGoing evaluation. The fingerprint extractor and matcher were ranked as the most accurate for both FV-STD-1.0 and FV-HARD-1.0 benchmarks.
In 2019 MegaMatcher palm print matching algorithm has shown the top result at the FVC-onGoing evaluation. The algorithm was the most accurate overall and fastest among the five most accurate matchers.

PFT II AND PFT III (PROPRIETARY FINGERPRINT TEMPLATE) EVALUATIONS

Different versions of Neurotechnology's fingerprint recognition algorithm were submitted to the NIST Proprietary Fingerprint Template Evaluation. The algorithm's template matching accuracy was among the best participants at the previous PFT II evaluation. Our latest submissions to the PFT II and the ongoing PFT III are in average the most accurate algorithms in all the experiments. See our comments for more information.

FpVTE (Fingerprint Vendor Technology Evaluations) by NIST

FpVTE 2012 – in 2015 NIST recognized Neurotechnology's fingerprint identification algorithm as one of the fastest and most accurate among the evaluation's participants.
FpVTE 2003 – one of the best reliability results in the Middle Scale Test were shown. Neurotechnology participated in FpVTE 2003 under the name Neurotechnologija.

FRVT Ongoing

In 2018 Neurotechnology has been ranked among 8 most accurate face recognition algorithm vendors out of 39, with tenth most accurate algorithm out of 78 in the FRVT leaderboard. The submission was also ranked as one of the best in two difficult scenarios, with second most accurate result on a complex dataset collected from operational photos related to ongoing criminal investigations, and fourth most accurate result with unconstrained, photojournalism-style photos.

FIVE (Face in Video Evaluation)

In 2015 Neurotechnology face recognition engine to the NIST Face in Video Evaluation (FIVE). In average the submitted algorithm was ranked among top 8 most accurate face recognition algorithms out of 16 vendors.

IREX Evaluations by NIST

IREX 10 – in 2020 Neurotechnology's iris recognition algorithm has been judged by NIST as the second most accurate among the IREX 10 participants. The submitted algorithm featured much faster template creation and search time, and much smaller template size than the only more accurate contender.
IREX IX – in 2018 Neurotechnology's iris recognition algorithm has been judged by the NIST as the second most accurate among the participants. The accelerated version of the algorithm was nearly 50 times faster than any other matcher in the NIST IREX IX evaluation.
IREX IV – in 2013 Neurotechnology's iris recognition algorithm has been judged the by the NIST as one of the fastest and most accurate among the participants.
IREX III – in 2012 MegaMatcher iris matching algorithm was the second fastest and provided 3 times higher recognition accuracy than the only faster contender.
On-Premise System Architecture and Components

Large-scale biometric projects may have specific system requirements. MegaMatcher Automated Biometric Identification System (ABIS) features modular architecture, which allows to adapt to different functional and performance requirements.

MegaMatcher ABIS 12.1 is available as on-premise solution, which is described on this page, and as cloud service.

MegaMatcher ABIS 12.1 on-premise architecture schema.
MegaMatcher ABIS 12.1 on-premise solution includes these services and applications:
MegaMatcher ABIS Client Application
Enrollment
Identity Management
Data Analysis
System administration
MegaMatcher ABIS API
Management Service
Image Processing Service
Matching Service

MegaMatcher ABIS Client Application

MegaMatcher ABIS Client Application is a modular web-based application designed to operate with the MegaMatcher ABIS solution. It provides the following functionalities:

Identity Management – provides the following functionalities:
  • Enrollment – registration of person's biometric and demographic data with the following features:
    • Multi-biometric capture – supports fingerprint, palmprint, face and iris biometric modalities. A wide range of biometric devices is supported, including more than 150 fingerprint readers and iris scanners models, as well as out-of-the-shelf web and IP cameras.
    • Demographic data – allows to specify name, sex, date of birth, address, as well as other data fields, which are unlimited and can be customized.
    • Live registration – person's biometric and demographic data are entered into the system in real time from stationary or mobile enrollment stations.
    • Data import – the system provides means to import a batch of biometric data from image files, which were captured in the field or taken from another system.
    • Quality check – evaluation of image quality based on standard and proprietary algorithms
    • ICAO compliant facial image capturing – checks image pixelation, washed out colors, background uniformity, face darkness, skin tone, skin reflections, glasses reflections, red eyes and looking away eyes. The red eyes can be corrected automatically.
  • Update – personal records can be refilled or changed at any time with new biometric and demographic data.
  • Delete – personal records, which are outdated or incorrect, can be deleted
  • Demographic and biometric search – allows to find particular person not only by name or surname, but also by face, fingerprint, palmprint and iris biometric records.
Adjudication – a visual tool designed for operators to examine questionable biometric matching results, which are forwarded by the Matching Service, and take appropriate decisions. The questionable matching results, like non-matching templates or detected duplicate(s), are determined by defined restrictions in the System administration module. The tool provides easy examination of given biometric data by providing detailed templates review.
  • Fingerprint, palmprint, face and iris images are shown in side by side or overlapping view with the possibility to transform and enhance biometric images so that some aspects could be seen more clearly.
  • Modality-specific features (such as fingerprint minutiae, cores, deltas).
  • Matching scores for each biometric modality.
  • Problematic fingers.
  • Matched data which is highlighted for easier inspection (such as matched minutiae are optionally connected by lines for visualization).
  • Age estimation and gender classification are performed automatically to help human operators in resolving questionable cases.
Latent Fingerprint Editor – a visual tool which provides ability to analyze the fingerprint image, perform automatic and manual enhancements as well as fingerprint feature points management. It is designed for the cases, when automated image processing is unable to extract all minutiae or extracts a lot of false minutiae from a latent fingerprint image (for example, taken from the crime scene). Therefore, an expert should manually edit a fingerprint template in order to submit it to the system for the identification.
System administration – designed to manage and monitor MegaMatcher ABIS system using role-based access control and consists of the following parts:
  • System Management – provides ability to manage MegaMatcher ABIS system's state.
  • System Monitoring – dedicated for visual MegaMatcher ABIS real-time execution monitoring.
  • User Management – dedicated to create and revoke users as well as provide or restrict access permissions to particular MegaMatcher ABIS modules.
  • System Configuration – provides ability to define criteria for search engine (such as matching thresholds for identification and verification).
  • Report Generation – statistical reports with filtering possibility are provided for full MegaMatcher ABIS actions inspection.

MegaMatcher ABIS Client Application works on Microsoft Windows, Linux, Android and iOS.

Enroll main
MegaMatcher ABIS client screenshot - enroll mainClick to zoom

Enroll fingerprints
MegaMatcher ABIS client screenshot - enroll fingerprintsClick to zoom

Confirm enrollment
MegaMatcher ABIS client screenshot - confirm enrollmentClick to zoom

Adjudication main
MegaMatcher ABIS client screenshot - adjudication mainClick to zoom

Fingerprints adjudication
MegaMatcher ABIS client screenshot - fingerprints adjudicationClick to zoom

Main dashboard
MegaMatcher ABIS client screenshot - main dashboardClick to zoom

MegaMatcher ABIS API

MegaMatcher ABIS API is a web services-based interface (RESTful) designed for easy and quick integration with third-party systems. It provides all necessary functionalities including identity management, adjudication and system administration. Integrators can develop client-side applications using the MegaMatcher ABIS API to meet specific requirements.

Management Service

MegaMatcher ABIS is designed to maintain system's integrity ensuring that individual operations are executed consistently. The Management Service is the core component of MegaMatcher ABIS responsible for orchestrating the work of all system services and managing the workflow of operations.

  • Transaction management – each operation forms a transaction which is passed through several stages:
    • Validation – ensures the sufficiency of provided biometric and demographic data.
    • Data preparation – conducts biometric data quality analysis, adjustments and biometric template extraction.
    • Matching – conducts biometric comparison using predefined parameters.
    • Adjudication – manual inspection of matching results.
  • Workflow management – the ability to adjust operation workflow to meet particular business needs.
  • Security – role-based access control ensures that user without privileges will not be able to execute a particular operation.

Image Processing Service

The Image Processing Service is based on MegaMatcher 12.1 biometric template extraction technology. This component is used for biometric data quality assessment and biometric data extraction from fingerprint, palmprint, face and iris images.

These functions are performed by the Image Processing Service:
  • Quality assessment – images with fingerprints, palmprints, faces and irises are processed by proprietary and standard algorithms to determine their quality and suitability for biometric operations.
  • Fraud assessment – detection of non-human/synthetic biometric, e.g. finger liveness and presence of eye contact lens.
  • Compliance assessment – determination of fingerprint misplacement, facial occlusions, ICAO compliance etc.
  • Feature extraction – creation of proprietary biometric templates from the given fingerprint, palmprint, face and iris images.
  • Facial attributes detection – automatic age estimation and gender classification can be performed for a given facial image.

Matching Service

The Matching Service is a high-performance biometric engine which performs the following operations:
  • Identity verification – matching a biometric template against a selected database record (1-to-1).
  • Person identification – matching a biometric template against the whole database (1-to-many).
  • Records deduplication – matching all biometric templates in the database against each other (many-to-many). The questionable matching results are resolved via the adjudication procedure, either automatically or by human operators.

The service must be added to MegaMatcher Accelerator 12.1 and it allows the engine to execute ABIS specific functionalities.

The Matching Service is a scalable unit controlled by the Management Service. Multiple units can be arranged into a cluster to meet performance and capacity requirements for a specific project . Each particular Matching Service unit requires to be connected to the corresponding MegaMatcher Accelerator unit.

Customization

Taking into account different business needs and specific project requirements, some customization might be required. This may include developing custom user interface, workflow scenarios, person's record structure (biometric and demographic data, additional data fields).

Neurotechnology offers custom project development service for providing a customized MegaMatcher ABIS based solution, as well as implementing additional functionality.

System Requirements

MegaMatcher ABIS software is distributed as a set of Docker containers.

Any modern Linux distribution can be used as a base operating system, including Debian, Ubuntu or CentOS.
The distribution package includes scripts for automatic installation of all necessary components.
Each software component can run in a physical or virtual environment.
Below are minimal requirements for each component:
Matching Service – depending on the edition, each node should be installed on hardware with the following specifications:
  • Development Edition and Standard Edition:
    • PC or server hardware with at least Intel Core i7-4771 processor (4 cores, 3.5 GHz) or equivalent;
    • 16 GB of RAM.
  • Extended Edition needs the same hardware as MegaMatcher Accelerator Extended, including:
    • HPE ProLiant DL360 Gen10 Server.
    • 2 x Intel Xeon Gold 6126 processor (12 cores, 19.25M cache, 2.6 GHz) or equivalent hardware.
    • RAM, depending on used biometric modalities:
      • 128 GB RAM for single modality;
      • 256 GB RAM for two modalities;
      • 384 GB RAM for three modalities;
      • 512 GB RAM for four modalities.
    • 6 x 300 GB HDD or SSD with similar capacity.
  • Extreme Edition needs the same hardware as MegaMatcher Accelerator Extreme, including:
    • HPE ProLiant DL360 Gen10 Server.
    • 2 x Intel Xeon Gold 6126 processor (12 cores, 19.25M cache, 2.6 GHz) or equivalent hardware.
    • RAM, depending on used biometric modalities:
      • 512 GB RAM for one or two modalities;
      • 1024 GB RAM for three modalities.
    • 2 x 400 GB SSD or HDD.
Management Service and Administration Service can be run on the same machine with the following specifications:
  • Server hardware with Intel Xeon Gold 6126 processor (12 cores, 19.25M cache, 2.6 GHz) or equivalent hardware.
  • At least 32 GB of RAM.
  • An application server to run the Management Service.
  • A web server for accessing the Administration Service.
MegaMatcher ABIS Client Application can be run on the following platforms:
  • PC or tablet, running Microsoft Windows OS or Linux
  • Mac, running macOS
  • Smartphone or tablet running Android 5.0 or later
  • iPhone or iPad running iOS
Cloud Service

MegaMatcher Automated Biometric Identification System (ABIS) cloud service is designed to make the system more accessible from multiple platforms and locations. The 30-day trial and subscription for the service are available at the MegaMatcher.online website.

The cloud service is based on the same MegaMatcher ABIS technology, which is used for on-premise systems. There is no need to purchase any hardware, as all resource-demanding tasks are performed on the cloud. In addition to the general advantages of using cloud architecture, the service features:

Flexible resources – enough data storage space and processing power can be offered based on the current project requirements. The subscription-based licensing allows to choose the required system performance for reaching project goals, while paying a reasonable price for the services. Such flexibility is especially useful for short-time tasks like database deduplication, as well as scalable systems.
Easy integration – web service-based RESTful API is designed for easy and quick integration with third-party systems. Additionally, Java and .NET SDK libraries are provided.
Security – the cloud service uses role-based access control with industry-standard authentication mechanisms. System activities are assessable for audit.
For a price please call on
+1-202-681-8115
or
Send us an email on
uia.sales@unifyia.com