Abstract:
Biometrics is used for human recognition which
consists of identification and verification. In an
identification application, the biometric device reads
a sample and compares that sample against every
record or template in the database. Identification
applications are common when the goal is to identify
criminals, terrorists, or other particularly through
surveillance. Personal face recognition is crucial for
applications such as access control, smart card
verification, surveillance, human-computer
interaction, etc. Also, faces are integral to human
interaction. Manual facial recognition is already
used in everyday authentication applications.
In this paper, a novel subspace method is
proposed for face recognition. A new face
recognition method DiaPCA is based on PCA
(principal Component Analysis) and KNN (Kth
nearest neighbor classifier). The recognition process
consists of three stages: preprocessing, dimension
reduction by using PCA, and matching of the
extracted feature using KNN. Combination of
DiaPCA and KNN is used for improving the
capability of PCA when a few samples of images are
available. In contrast to standard PCA, DiaPCA
directly seeks the optimal projective vectors from
diagonal face images without image-to-vector
transformation. DiaPCA reserves the correlations
between variations of rows and those of columns of
images. DiaPCA is much more accurate than PCA.
The motivation of this research is to provide the
personal identification from the National
Registration Card (NRC card).
