Using a biometric imaging and identification system called PhotoFace developed at the University of the West of England (Bristol, England) and Imperial College London (London, England), researchers from the University of Bath (Bath, England) have discovered that nose-shape scanning has a comparable recognition rate to that achieved by applying conventional face-recognition methods to the nose region, with a faster computation rate.
In addition to scanning the 3-D shape of volunteers' noses and classifying them as either Roman, Greek, Nubian, Hawk, Snub, or Turn-up (the six main nose shapes) alone, the researchers analyzed the ridge profile, the nose tip, and the nasion area (the concave section of the nose between the eyes). From this data, the ratio of the nose-tip and nasion widths to the ridge length and the ridge curvature were compared against a database of 36 people, with identification results comparable to other biometric techniques applied to the nose region. Because noses are more difficult to conceal than eyes and change little as facial expressions change, nose-shape biometrics could improve human-threat identification in comparison to iris or whole-face biometric methods. Contact Adrian Evans at [email protected].