Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Networks

Biometrics systems now recognize multiple traits

Posted: 01 Jun 2001 ?? ?Print Version ?Bookmark and Share

Keywords:biometrics? fingerprinting? multimodal biometrics? iris scanner?

Many companies are turning to biometrics to truly recognize authorized users through unique, non-transferable and non-duplicatable traits, such as a face, voice or fingerprint, says Michael Ruehle.

Michael Ruehle is senior vice president of marketing at Bioid America, the US subsidiary DCS AG.
As companies increase their use of network communications and Internet transactions, the safeguarding of sensitive information becomes a vital issue. Conventional authentication solutions are based on transferable knowledge (passwords) or possession (smart cards) that can be obtained by unauthorized users. Many companies turn to biometric solutions to truly recognize authorized users through unique, non-transferable and non-duplicatable traits, such as a face, voice or fingerprint, to offer corporate networks and offices maximum protection. According to the International Biometrics Group, the global biometric market will generate $1.1 billion in revenue in 2003, up from $340 million in 2000.

But early adoption has been hindered by the high costs of early products, coupled with complex installation and integration with existing security systems. In addition, many biometric solutions are monomodal, meaning they analyze only one human trait. As a result, their application tends to be limited because they have low tolerance to variation. Should a user wear glasses one day instead of contact lenses, or develop a minor finger injury, recognition problems could arise.

To counteract those challenges, multimodal biometric solutions have been developed that recognize more than one trait, helping to ensure tolerance, high recognition accuracy and maximum access protection. For example, software that identifies face, voice and lip movement allows for a higher degree of tolerance to variation. If a user has a sore throat and the voice pattern cannot be recognized, the software falls back on the two visual traits for verification.

Ensuring that security is not compromised, certain multimodal solutions give the system administrator the option of weighing the importance of each identifiable trait, customizing to decide which trait, if any, should be given priority. A balance can be struck between increased tolerance for slight changes in personal traits, leading to a lower false-rejection rate, or requiring agreement of all traits, resulting in a lower false-acceptance rate.

The majority of multimodal solutions have been layeredthat is, trait recognition is done sequentially. This process tends to be slow. A simultaneous multimodal technology has been developed that can identify several human traits at once for a quicker, less obtrusive process.

Multimodal biometrics also relieves the technology of its perceived obtrusiveness. Recognition can be as simple and natural as saying hello to a friend. For example, the simultaneous identification of face, voice and lip movement from a single video clip eliminates the need for physical contact, such as placing a hand onto a plate for hand geometry identification. And, the client need not make a huge investment in specialized biometric devices like an iris scanner. Instead, a standard web camera and microphone can be sufficient.

This technology is poised to alter the biometrics landscape and increase its acceptance by increasing security, simplifying use and lowering costs.

Article Comments - Biometrics systems now recognize mul...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top