Software Sniffs out Criminals by Their Nose
Forget iris and fingerprint – scanning nose could be a faster and easier to verify the identity of a person, according to scientists at the University of Bath.
With concerns about illegal immigration and identity theft, authorities are increasingly using the physical characteristics of an individual, known as biometrics, to confirm his identity.
Unlike other facial features used for biometric data, such as eyes or ears, nose are difficult to hide and does not change much by facial expression.
Dr. Adrian Evans and Adrian Moorhouse, University Department of Electrical and Electronic Engineering, decided to investigate whether the images of the noses of people could use to recognize individuals.
We used a photographic system called PhotoFace (http://www.uwe.ac.uk/cems/research/groups/mvl/index.shtml), developed by researchers at the University of West England in Bristol, to explore the 3D shape of the nose of the volunteers and the software used to analyze according to six main forms of the nose: Roman, Greek, Nubian, Hawk, blunt and Turn-up.
Instead of using the whole shape of the nose, the researchers used three characteristics in their analysis: the profile of the ridge, the tip of the nose and nasal root or the section between the eyes at the top of the nose .
They combined the curvature of the ridge with the proportions of the tip and the nasion length width and crest. The combined ratio was used to distinguish between a database of 36 persons.
While the researchers used a relatively small sample, they found that the nose scan showed good potential for use as biometric data, with a rate of good recognition and image processing rate faster than the biometric techniques as recognizing the face.
Dr. Evans said: “Noses facial features are prominent, and yet its use as a biometric has been largely unexplored. We wanted to know how well it could be in recognition of individuals in a database.
“There is no magic biometrics – iris biometrics are powerful, but can be difficult to capture with precision and can be obscured by eyelids or glasses.
“Noses, however, are much easier to photograph and are more difficult to hide, so a system that recognizes the nose work better with a theme uncooperative or for covert surveillance.
“We’ve only tried this on a small sample of people, but the technique clearly demonstrates the potential perhaps to be used in combination with other identification techniques.
Professor Melvyn Smith led the team from the University of West of England (UWE), who developed the system PhotoFace.
He said: “This collaborative project with bath is a very interesting job with great potential. PhotoFace an innovative 3D face data capture system developed as part of a project funded by EPSRC participation UWE, Imperial College, the Ministry Interior (Development Section of Science) and General Dynamics Corp.
“It works by taking pictures illuminated by a beam from different angles so that four images are taken at a very rapid succession of each point of the face, each under different conditions of controlled illumination.
“The technique is known as photometric stereo and Machine Vision Laboratory UWE is one of three centers in the UK with experience in this field. The software works on color, surface orientation and depth of each point of the face by analysis of the shadows within each of the photos.
“The technique is capable of achieving a level of detail is beyond the current competing technologies and can be extended to a myriad of other applications ranging from industrial surface inspection to cosmetics.
The researchers plan in the future to build a large database of the nose to test and refine the software to see if you can select people from a larger group of people, or dividing the family of the same family.