Make way for knobbly kneed ID

Knees and ears are perhaps unlikely measures of identity, but they could soon rival fingerprints as personal identification. Rawlson King reports

Traditional biometrics, such as DNA and fingerprints, are well known and are widely used due to declining prices, technology miniaturisation and software optimisation. However, new developments are emerging that measure less conventional biometrics for personal identification.

In the United States, the Lawrence Technological University at Michigan last month announced that one of its researchers had created a biometric identification system which analyses a person’s knees. Lior Shamir, a computer scientist at the university, has developed a system, based on MRI, which can quickly register and identify people in a crowd. Dr Shamir’s tests looked at knee scans from approximately 2,700 people and reportedly achieved an accuracy rate of around 93 per cent.

Although this rate is a relatively low score for a standalone biometric, because of the difficulty to deceive this measure, it could be coupled well with other biometrics such as facial recognition or fingerprints to verify identity.

Mark S. Nixon, a computer scientist at Southampton University, has developed an ear biometric. Using photos of individual ears matched against a comparative database, Professor Nixon argues that he has developed a means of non-invasive identification that is as distinctive as fingerprints. He believes ear biometrics is a superior measure due to the fact that ears remain relatively the same, despite the ageing process.

Gait recognition could be applied in law enforcement to identify criminals who cover their faces

These elements rely on someone being relatively static when they are being scanned. Biometric recognition systems have also been developed based on gait, the measure of how an individual walks. New gait technology has been developed by the UK’s National Physical Laboratory (NPL) in collaboration with the Centre for Advanced Software Technology (CAST) and BAE Systems. NPL used closed circuit television (CCTV) images from its headquarters to record a person’s specific walk. The system works by extracting someone’s silhouette from the background, utilising software and recording the patterns in order to identify the person.

NPL is not the only organisation that is conducting gait recognition research. Also involved are Professor Martin Hoffman at the Technical University of Munich and Professor Daigo Muramatsu at Osaka University. Potential uses are many; it could be applied in law enforcement to identify criminals who cover their faces or to improve security in highly sensitive environments such as airports.

Airports typically use well-known iris recognition technologies for authentication of visa applicants at passport control checkpoints. These systems are only concerned with verifying identity against a static image of an eye.

A research team at the University of Tampere in Finland, however, has proposed creating a biometric security system based on eye movement. Through video surveillance, Professor Martti Juhola and colleagues have proposed a system that would monitor saccades, the unique and rapid involuntary eye movements that all people make. Saccades are quick, simultaneous movements of both eyes in the same direction and once saccades are underway, they cannot be altered by will. A saccade is also an involuntary consequence of turning the head to one side or a sudden motion detected in one’s visual periphery.

Saccades are probably the simplest eye movements to detect with signal analysis, according to the research team. They are the fastest eye movements and very easy to trigger by asking an individual to look at one target, and then another on a computer screen.

The researchers note that it would be much more difficult to hack or spoof an individual’s pattern of saccades than to emulate their iris with contact lenses or their fingerprints with patterned silicone pads, or by way of other forged images or prosthetics, because of the uniqueness of eye movements.

Preliminary tests, reported in the International Journal of Biometrics, indicate that a biometric identity verification could be undertaken in as little as 30 seconds with 30 to 40 saccades being recorded, providing an accuracy rate of 90 to 100 per cent.

A researcher at Texas State University has also recently developed and patented a new authentication method based on scanning the oculomotor nerve in the eye.

In terms of other emerging technologies, the Biometrics Research Group, a Toronto-based leading market research supplier and consultancy to the global biometric industry, projects that healthcare applications will also be a serious driver for new emerging biometric identifiers.

Within the next few years, we can expect new biometrics that aid health and medicine by measuring the foot and other biorhythms. For example, US researchers at Carnegie Mellon University are currently turning to human feet as a new biometric identifier. Receiving $1.5 million a year, the university’s Pedo-Biometrics Research and Identity Automation Lab will team up with a Canadian company, Autonomous ID, to test a new identity automation technology using the human foot.

Research will include identification, security and health applications, such as detecting the onset of diseases including diabetes and Parkinson’s. Biometrics Research Group also projects that the increased use of fitness applications and consumer electronics will spawn new gesture-based biometrics.



Market researchers Biometrics Research Group, Inc. projects that unconventional biometrics will begin to constitute a larger portion of research dollars spent on R&D within the biometrics industry. They conservatively estimate that the US government spends at least $450 million per annum, in inflation-adjusted 2010 dollars, on pure biometric research. Much of this spending is driven by homeland security and the defence budget. The firm expects at least $10 million to $15 million of government biometric research spending to be directed to alternative biometric identifiers, such as gait or oculomotor nerve scanning.



An interesting new biorhythm-based biometric device is a smartphone-enabled cardiogram from a startup backed with venture capital from Qualcomm. It’s called the AliveCor, a skin that you add to your iPod which turns it into a fully-functional electrocardiogram or ECG heart monitor. The prototype device has actually been used to diagnose a heart attack and save a life on a plane, and was recently cleared by the Food and Drug Administration for sale in the United States for use by licensed medical professionals to record, display, store, transfer and evaluate single-channel ECG rhythms. AliveCor plans to sell the device to both consumer and medical markets, and has received permission to do so in the European Union.



Researchers at the Polytechnic Institute of New York University are working on developing gesture-based authentication for tablet computing that one day might replace the use of passwords. Computer scientists at the university are working on new ways to recognise a user’s biometric traits, such as the unique shape of a hand or finger, in order to replace passwords that can be easily compromised. Studies found that tablet apps with gesture-based authentication are more secure, memorable and enjoyable alternatives to pass codes and passwords.