Abstract
DESIGN OF A BIOMETRIC DATABASE SYSTEM USING A SECURED NETWORK INTEGRATION
Idigo V. E., Onwujei Augustine I.*, Okezie C. C., and Okafor K. C.
ABSTRACT
This work used composite (hybrid) Principal component analysis (C-PCA) as a statistical procedure that uses an orthogonal transformation to convert a set of image observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. In this case, the number of principal components is less than or equal to the number of original variables. This transformation concept was applied in novel facial recognition biometric attendance register for accurate time monitoring and impersonation detection incidences. Emgu camera video dynamic link library (Emgu.CV.dll) and MySQLite-3 were used in deriving the PCA Eigen vectors. After deploying the system, various enrolments were carried out. Eigen verifications were also followed up to check for the image pattern matching. Verified images are certified while unverified images are flagged as illegal image templates. Also, the network deployment context was achieved based on carley graph Ethernet LAN environment. In this case, communication from the IP based cameras are relayed to secured database servers. This work showed the merits and application domain while concluding that with principal components image transformation, facial recognition systems can be used in complex surveillance systems.
[Full Text Article] [Download Certificate]