Biometric scientists at the University of Southampton say they can identify ears with a 100% success rate.
In a new paper, scientists from the University’s School of Electronics and Computer Science (ECS) described how a technique called the image ray transform can highlight tubular structures such as ears, making it possible to identify them.
The research describes how the transform is capable of highlighting tubular structures such as the helix of the ear and spectacle frames and, by exploiting the elliptical shape of the helix, can be used as the basis of a method for enrolment for ear biometrics.
Professor Mark Nixon, one of the UK's earliest researchers in this field, first used ears as a viable biometric back in 1999. Back then, he said that ears have certain advantages over more established biometrics as they have a rich and stable structure that is preserved from birth to old age and instead of aging they just get bigger. The ear also does not suffer from changes in facial expression and it is firmly fixed in the middle of the side of the head against a predictable background, unlike face recognition which usually requires the face to be captured against a controlled background.
But ears can be concealed by hair, so they had to come up with new algorithms to make it possible to identify and isolate the ear from the head.
The technique presented by the scientists achieves 99.6% success at enrollment across 252 images of the XM2VTS database, displaying a resistance to confusion with hair and spectacles. These results show great potential for enhancing the detection of structural features.
“Feature recognition is one of the biggest challenges of computer vision,” said Alastair Cummings, the PhD student for the research. “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.”
IEEE Fourth Conference on Biometrics: Theory, Applications and Systems, September 2010, Washington DC, USA. A Novel Ray Analogy for Enrolment of Ear Biometrics by professor Mark Nixon, Dr. John Carter and Alastair Cummings