Abstract :
— This paper proposes to implement the biometric
security system spoofing detection on palm vein, face and
iris image patterns. A Biometric system is essentially a
pattern recognition system that makes use of bio-metric
traits to recognize individuals. In a biometric security
system, the data hiding approach is involved to conceal the
secret personal informatics for enhancing the privacy
protection. There are negative effects on recognition
performance on fingerprint and palm print biometrics due to
the some factors like the dryness or dirt in the finger and it
also varies with age, for instance, this system is not
appropriate for children, because the size of their fingerprint
changes quickly. This paper depicts a proposed method
which avoids the above negative factors. Palm vein, face
and Iris patterns stand out from the host of intrinsic
biometric traits for the development of a recognition system
that can meet all the security expectations of a biometric
system. In the proposed method palm vein, face and iris
image patterns spoofing can be easily detected using Neural
Network (NN) with the help of GLCM properties. Vein
patterns are the network structure of blood vessels
underneath the human skin that are almost invisible to the
naked eye under natural lighting conditions and can be
acquired only when employing infrared illumination. The
texture of the blood vessels and Iris of different individuals
has been proven to be distinctive even among identical
twins. The selected image of palm veins, face and Iris is
aligned and cropped according to the key points. The image
is enhanced and resized. The features of palm vein, iris and
face are compared with database image feature vectors and
are recognized using Probabilistic Neural Network
classifier (PNN). Finally the performance of multimodal
system along with stenographic approach will be measured
with accuracy and it proves to provide better matching rate
than earlier approaches.
Keyword :
Biometric, Artificial Neural Network, Probabilistic Neural Network Classifier, Gray-Level Concurrence Matrix, Non-Subsampled Contourlet Transform