Follow the Nose, It Always Knows
March 3, 2010
You know Toucan Sam of Fruit Loops fame might have been on to something years ago when he coined the famous phrase, “Follow the nose, it always knows!” Scientists from the University of Bath in the UK are using the most prominent feature on an individual’s face, their nose, as a way to confirm their identity through facial recognition software.
Of course the author of the article, Software Sniffs Out Criminals by the Shape of Their Nose, didn’t do his work justice by conjuring up images of some Aryan Brotherhood experiment with his title. However, if you look further, it’s an interesting concept. Is this setting the stage for background checks and employment screening being conducted by looking at a person’s schnozola?
With worries about illegal immigration and identity theft, authorities are increasingly looking to using an individual’s physical characteristics, known as biometrics, to confirm their identity.
Unlike other facial features used for biometrics, such as eyes or ears, noses are difficult to conceal and also aren’t changed much by facial expression.
Dr Adrian Evans and Adrian Moorhouse, from the University’s Department of Electronic & Electrical Engineering, decided to investigate whether images of people’s noses could be used to recognise individuals.
They used a photographic system called PhotoFace (http://www.uwe.ac.uk/cems/research/groups/mvl/index.shtml), developed by researchers at the University of the West of England in Bristol, to scan the 3D shape of volunteers’ noses and used computer software to analyse them according to six main nose shapes: Roman, Greek, Nubian, Hawk, Snub and Turn-up.
Instead of using the whole shape of the nose, the researchers used three characteristics in their analysis: the ridge profile, the nose tip, and the nasion or section between the eyes at the top of the nose.
They combined the curvature of the ridge with the ratios of the tip and nasion widths and ridge length. This combined ratio was then used to distinguish between a database of 36 people.