Biometrics more accurate, but uptake 'disappointing'


Biometrics technology has become more accurate of late, mainly because it is factoring in multiple body parts besides just a person’s fingerprints to make the identification. However, one analyst says its development from a security and use case standpoint is disappointing, according to a ZDNet report.

Kris Ranganath, director of technology and solutions of NEC, pointed out that the latest developments in biometrics focus on multimodal fusion matching, or “person centric identification”. This means that any available biometric data generated by a person can be used for verification. This, he noted, is unlike the past when tools depend solely on a single mode of a person’s biometrics such as fingerprints.

Today, a person’s face, iris, the veins in his finger or palm, and even body odor can be used for identification purposes, the executive said. By combining these metrics, overall accuracy for biometrics has been improved and false rejection rates and false acceptance rates reduced.

This, in turn, will help better prevent fraud and improve user experience, Ranganath stated.

Already, multimodal based products are being used. For instance, India’s unified identification (UID) project Aadhaar, which was rolled out in 2010, aims to gather citizens’ fingerprints and iris data for authentication purposes, the NEC director pointed out.

Beyond multimodal fusion matching, biometrics is also being used as part of multi-factor authentication processes, said Lee Wei Jin, Southeast Asia director of sales in identity and access management at security company HID Global.

In banking, specifically, such offerings can help address basic fraud elements as well as extend banking services to reach the illiterate and barely literate rural masses, added Ricardos Khoury, regional vice president and head of banking at Wincor-Nixdorf Asia-Pacific.

Brunei’s Bank Islam Brunei Darussalam (BIBD) is one case study, having deployed biometric ATMs in November last year, he said. These machines enable customers to gain access to their accounts using their bank card and fingerprint, thus doing away with the need to remember a password.

Other aspects of biometrics such as voice recognition remain on the periphery of the IT industry, though.

Lee said voice-based technology had been in the spotlight in recent times due to the introduction of Apple’s voice recognition feature, Siri. It uses a natural language user interface to answer questions, make recommendations and perform actions such as make a phone call upon the user’s voice command, he said.

However, the adoption of voice identification tools remains low because voice as a metric is highly dependent on emotional factors and a person’s state of physical health, the executive said. For example, a person having a sore throat or bronchitis will affect the accuracy of the service.  Similarly, external factors such as traffic sounds and other ambient noises will also impact the level of recognition, he said.


Please enter your comment!
Please enter your name here