An Introduction to Biometrics - Multimodal Biometric Systems


What are mulitmodal biometric systems?
A biometric system that utilizes more than one core technology for user authentication is referred to as multimodal (in contrast to monomodal). There are three types of multimodality in the biometric world:

Synchronous multimodality involves the use of multiple biometric technologies in a single authentication process. For example, biometric systems exist which use face and voice simultaneously, reducing the likelihood of fraud and reducing the time needed to verify. Systems that offer synchronous multimodality can be difficult to learn, as one must interact with multiple technologies simultaneously.

Asynchronous multimodality describes systems that require that a user verify through more than one biometric in sequence. Asynchronous multimodal solutions are comprised of one, two, or three distinct authentication processes. A typical user interaction will consist of a verification on finger scan, then face if finger is successful. The advantage of added security – it is highly unlikely that a user will break two systems – is offset by a reduction in convenience. In addition to the time required to execute these separate submissions correctly (such verification can require 10 seconds of submission) the user must learn multiple biometric processes, as in either/or systems. This can be a challenge for both physical and logical access scenarios.

Either/or multimodality describes systems that offer multiple biometric technologies, but only require verification through a single technology. For example, an authentication infrastructure might support facial, voice, and fingerprint at each desktop and allow users to verify through any of these methods. A number of vendors have developed enabling middleware that allows for authentication by means of various biometrics. The benefit of this system is that biometrics, instead of passwords, can be used as a fallback. To have access to either/or multimodality, a user must enroll in each technology. To use finger, face, and voice, for example, one must become familiar with three devices and three submission processes. A key performance indicator in biometrics is ease-of-use, requiring familiarity with multiple processes can be problematic.


Trials
Human Recognition Systems, the only independent multi-biometric systems integrator in the UK, have just commenced a trial of biometric technology for staff access control, officially sponsored by Manchester Airport and the Department for Transport. The trial is aimed at assessing the practical use and user acceptance levels of biometric technology specifically in an airport environment. The technologies on trial include iris recognition from Panasonic and hand geometry recognition from Recognition Systems Inc. The two systems are running in conjunction with a photo ID system which presents a photo of an individual every time they present themselves to a biometric reader. A system administrator then manually checks that the individual’s photo is correct and records a valid match. Both manual and system logs record several data sets including false accept rates, false reject rates and transaction speed times. Manned for 14 hours every day, for a two month period, the trial is expected to amass some 500 participants. Pre and post trial user questionnaires will also provide an indication of user acceptance.

The BioSec project have produced 2 biometric security scenarios: physical access and remote access. The physical access scenario is implemented in Helsinki-Vantaa Airport, Finland and the remote asccess scenario is implemented in Spain and Portugal. Scenarios in the 2nd year will employ 3D face recognition and a multimodal system with fingerprint and face recognition. It's planned that these are implemented in the physical access security scenario. The remote security scenario will possibly cover a multimodal system with fingerprint and iris, and additionally voice recognition.

The Japanese Ministry of Land, Infrastructure and Transport (MLIT) and the JAL Group will carry out tests of secure and smooth flight airport procedures at New Tokyo International Airport in Narita. The tests will evaluate the practical use of biometrics, i.e. facial, iris and fingerprint authentication technology for check-in, security checks and immigration control.


Standards

Standards under development:

  • INCITS PN-1627-S: Information Technology-Evaluating Multi-Modal Biometrics Systems: Concepts of Operation and Methods of Performance Evaluation (study project)
  • ISO/IEC WD 24722: Multi-Modal Biometric Fusion


Further information


Manufacturers/suppliers of multimodal biometric technologies:

Human Recognition Systems
HumanScan AG
Identix Inc
LG Electronics Inc
NextgenID Ltd
Precise Biometrics
QVoice
Saflink


Acknowledgement

The information contained in this section was collected from the following source:

 

Last updated: 20.11.2009   © Copyright reserved    Website design: Digital Accessibility Team