Multimodal biometrics

Access & Identity Management Handbook 2006 Access Control & Identity Management

Biometrics measure an individual’s unique physical or behavioural characteristics, as a means to recognise or authenticate their identity.

Common physical biometrics includes fingerprints; hand or palm geometry; and retina, iris, or facial characteristics, whereas behavioural characteristics include signature, voice (which also has a physical component), keystroke pattern, and gait. While some technologies have gained more acceptance than others, it is beyond doubt that in the field of access control, biometrics has gained a measure of acceptance.

Biometric products provide improved security over traditional electronic access control methods such as RFID tags, electronic keypads and some mechanical locks. They ensure that the authorised user is present in order for access to take place. The user's authorised card or password pin cannot be stolen or lost to gain access.

The use of multimodal biometrics

In order for biometrics to be ultra-secure and to provide more-than-average accuracy, more than one form of biometric identification is required. Hence the need arises for the use of multimodal biometrics. This uses a combination of different biometric recognition technologies. In certain situations, the user might find one form of biometric identification is not exact enough for identification. This can be the case with fingerprints, where at least 10% of the population have worn, cut or unrecognisable prints. Multimodal biometric technology uses more than one biometric identifier to compare the identity of the person. Therefore, in the case of a system using say three technologies, ie, face, mimic and voice. If one of the technologies is unable to identify, the system can still use the other two to accurately identify against. Multimodal technologies have been in use commercially since 1998.

1:1 and 1:N matching

A biometric recognition system can be used in two different modes: identification (1:N matching) or verification (1:1 matching).

Identification is the process of trying to find out a person's identity by comparing the person who is present against a biometric pattern/template database. The system would have been pre-programmed with the biometric pattern or template of multiple individuals. During the enrolment stage, a biometric template would have been processed, stored and encrypted for each individual.

A pattern/template that is going to be identified is matched against every known template, yielding either a score or a distance describing the similarity between the pattern and the template. The system assigns the pattern to the person with the most similar biometric template. To prevent impostor patterns (in this case all patterns of persons not known by the system) from being correctly identified, the similarity has to exceed a certain level. If this level is not reached, the pattern is rejected.

With verification, a person's identity is known and therefore claimed a priority to search against. The pattern that is being verified is compared with the person's individual template only. Similar to identification, it is checked whether the similarity between pattern and template is sufficient enough to provide access to the secured system or area.

Multimodal biometrics in terms of FAR and FRR

Biometric systems use scores (also called weights) to express the similarity between a pattern and a biometric template. The higher the score, the higher the similarity is between them. As described in the previous section, access to the system is granted only if the score for an authorised individual (identification) or the person that the pattern is verified against (verification) is higher than a certain threshold.

In theory, authorised user scores (scores of patterns from persons known by the system) should always be higher than the scores of impostors. If this would be true, a single threshold, that separates the two groups of scores, could be used to differ between clients and impostors. This unfortunately is not the reality for real world biometric systems. In some cases, impostor patterns can generate scores that are higher than the scores of an authorised user's patterns (FAR or false acceptance rate). For this reason it is a fact that, however, the classification threshold is chosen, some classification errors may occur.

For example, you may configure the threshold with a high setting, which will reject all impostor patterns that exceed this limit. As a result, no patterns are falsely accepted by the system. But on the other hand the authorised user patterns with scores lower than the highest impostor scores are also falsely rejected. The opposite scenario would be to configure a low threshold that ensures no client patterns are falsely rejected. However, this would then allow a certain percentage of impostor patterns to be falsely accepted.

If you chose the threshold somewhere between those two points, both false rejections and false acceptances occur. This creates an access control environment which is obviously not ideal for high security installations.

Why multimodal?

By using more than one means of biometric identification, the multimodal biometric identifier can retain high threshold recognition settings. The system administrator can then decide the level of security he/she requires. For a high security site, they might require all three biometric identifiers to recognise the person or for a lower recognition site, only two out of the three. With this methodology, the probability of accepting an imposter is greatly reduced.

For more information contact Nathan Bearman, Eagle Biometrics, +27 (0) 21 423 4943,,

Share this article:
Share via emailShare via LinkedInPrint this page

Further reading:

Econz Wireless Timecard assists with employee screening
Issue 4 2020 , Access Control & Identity Management
Essential workers across a wide range of sectors are using Econz’ Timecard application to check employees’ current wellness status before starting work.

WizzPass COVID-19 screening solution
Issue 4 2020 , Access Control & Identity Management
WizzPass provides COVID-19 screening solution for both staff and visitors in one system.

Foot operated door opener
Issue 4 2020 , Access Control & Identity Management
Pro-Tech and Falcon Security have launched the new Foot Operated Door Opener (FODO), aimed at assisting in the creation of touchless ‘hands-free’ access control solutions.

Infection control access solutions
Issue 4 2020, Gunnebo Africa , Access Control & Identity Management
With many organisations planning to resume operations, the implementation of social distancing and infection control in these environments is a challenge.

Weatherproof Bluetooth padlock
Issue 4 2020, ASSA ABLOY South Africa , Access Control & Identity Management
The new ABLOY BEAT locking solution includes a novel digital key, a mobile app and an IP68-rated, weatherproof Bluetooth padlock.

Touch-free face recognition terminals
Issue 4 2020, Hikvision South Africa , Access Control & Identity Management
Hikvision announces touch-free MinMoe face recognition terminals for easier access control and time and attendance.

Contactless T&A
Issue 4 2020, LD Africa , Access Control & Identity Management
Matrix COSEC APTA is a mobile-based employee portal, allowing COSEC users to monitor and manage all their time and attendance, total working hours and leave options.

New touchless upgrades
Issue 4 2020 , Access Control & Identity Management
Invixium upgrades touchless biometrics technology and introduces fever detection enhancement kit.

Expanded entrance control
Issue 4 2020, Gunnebo Africa , Access Control & Identity Management
Gunnebo entrance control solutions are now available with body temperature control and building capacity management functionality.

Entrance control from Axis
Issue 4 2020, Axis Communications SA , Access Control & Identity Management
Access control solutions from Axis offer flexibility, agility and open platforms that are fit for both today’s needs and tomorrow’s.