Since the early 1990's face Recognition Technology (FRT) became an
active research area. Most of the current profile recognition algorithms
depend on the correct detection of fiducial points and the determination of
relationships among these fiducial points. Unfortunately, some features
such as concave nose, lips, flat chin, etc. make detection of such points
difficult and unreliable. Also the number and position of fiducial points
vary when pose changes even for the same person. In this paper, a
curvature-proposed technique is presented, which does not require the
extraction of all the fiducial points, but uses information contained in the
profile. The nearest neighbor interpolation method is used to smooth the
profile and then the curvature of the interpolated profile is computed.
Using the curvature coefficient values, the fiducial points, such as nasion,
chin, and forehead can be reliably extracted using a fast and simple
method. Then an Euclidean distance method is applied to match the face
profile based on the curvature coefficient values.
Experiments are performed on collected 50 clients with different ages, in
public area. Each had three shots differences in time capture and
illuminations, getting total of 150 images database, recognition rate of
96.67% and conclusion are presented and discussed.