Oct 11, 2010 09:53 GMT  ·  By

Biometrics scientists at the University of Southampton, found a way of identifying people by their ears, a method that has shown an almost 100% success rate.

Their technique is called the image ray transform and it can highlight tubular structures, just like ears, and specifically identify them.

The research carried out by Professor Mark Nixon, Dr John Carter and Alastair Cummings at University’s School of Electronics and Computer Science, focused on structures like the helix of the ear and spectacle frames, and found a way of enrolling ear biometrics.

This method reached a 99.6% success rate at enrollment across 252 images of the XM2VTS database, being very efficient against confusion from hair and spectacles.

Alastair Cummings, the PhD student for the research, said that “feature recognition is one of the biggest challenges of computer vision.

“The ray transform technique may also be appropriate for use in gait biometrics, as legs act as tubular features that the transform is adept at extracting.

“The transform could also be extended to work upon 3D images, both spatial and spatio-temporal, for 3D biometrics or object tracking.

“It is a general pre-processing technique for feature extraction in computer images, a technology which is now pervading manufacturing, surveillance and medical applications.”

Ever since 1999, Professor Nixon has carried out several studies that proved that ears could be used as biometrics, AlphaGalileo reports.

He said that ears have a stable structure that remains unchanged from birth to death, and instead of changing with age, they only get bigger.

Another positive thing is that the ears do not change because of facial expressions, and they remain fixed in the middle of the side of the head.

In spite of these arguments there was one issue that led Professor Nixon to carry out more research and develop a method of identifying and separating the ear from the head – the fact that hair can conceal the ears.

This new technique is described in a paper called A Novel Ray Analogy for Enrolment of Ear Biometrics, presented at the IEEE Fourth International Conference on Biometrics: Theory, Applications and Systems.