3D face recognition - Total security!
Biometric features are properties that characterise and therefore identify a person. Faces are undoubtedly the most important feature in interpersonal identification, a fact that is also corroborated by the integration of digital facial images into the new generation of passports. Other biometric features used for authentication and identification purposes include iris, retina, vein pattern, hand contour or DNA.
The systems presently available on the market, however, do not provide sufficient protection, especially for highly critical infrastructures. Different head orientations or unfavourable camera angles often lead to a substantial decrease of recognition performance and access authentication can be deceived relatively easily by means of photographs.
For this reason, Rittal presents, in cooperation with the Fraunhofer Institute IGD, an enhanced face recognition concept involving three-dimensional face data. 3D models allow the face image to be corrected in the case of head movement or unfavourable camera angles. In addition, these models are free from distortions, thus ensuring that basic dimensions, such as the distance between the eyes, are correctly preserved. The new system enables the comparison of unambiguous body features, making it practically tamper-proof.
After determining the 3D geometry and entering the corresponding text information, a computer is used to orient the face to a general frontal view and is subsequently compared with previously scanned 3D faces. Only sufficient concordance between the geometries will successfully authenticate the user and grant access to, for example, a data centre or a security room. Moreover, the system can be easily combined with further access controls in order to ensure an even higher level of security. 3D facial recognition completely links access to specific areas to a person's identity and is easy to log.
