Authors :
Ene Princewill Chigozie
Volume/Issue :
Volume 7 - 2022, Issue 8 - August
Google Scholar :
https://bit.ly/3IIfn9N
Scribd :
https://bit.ly/3TO9bnh
DOI :
https://doi.org/10.5281/zenodo.7055795
Abstract :
This research presents the development of
intelligent techniques for fingerprint and faces
recognition systems. This was achieved following data
collection, data acquisition, data processing, artificial
intelligence, training, and result presentation. The
intelligent technique was modeled using the structural
method to develop the algorithm for face and fingerprint
verification systems. The algorithms were implemented
with Simulink. The result showed that the average
Means Squre Error(MSE) for face is 4.7E-05Mu, that
for the fingerprint is 2.05E-05; the regression value for
face is 0.973 and 0.995 for the finger. The algorithm was
deployed as a face and fingerprint verification system
and the result were tested and validated using a tenfold
cross-validation approach. An accuracy of 98.6% was
achieved for face recognition and 98.87% was achieved
for fingerprint verification results. The performance was
compared with other algorithms and it was observed
that the new algorithm performs better.
Keywords :
Face Recognition, Fingerprint Verification, Training, Artificial Intelligence, Simulink, Artificial Neural Network (ANN).
This research presents the development of
intelligent techniques for fingerprint and faces
recognition systems. This was achieved following data
collection, data acquisition, data processing, artificial
intelligence, training, and result presentation. The
intelligent technique was modeled using the structural
method to develop the algorithm for face and fingerprint
verification systems. The algorithms were implemented
with Simulink. The result showed that the average
Means Squre Error(MSE) for face is 4.7E-05Mu, that
for the fingerprint is 2.05E-05; the regression value for
face is 0.973 and 0.995 for the finger. The algorithm was
deployed as a face and fingerprint verification system
and the result were tested and validated using a tenfold
cross-validation approach. An accuracy of 98.6% was
achieved for face recognition and 98.87% was achieved
for fingerprint verification results. The performance was
compared with other algorithms and it was observed
that the new algorithm performs better.
Keywords :
Face Recognition, Fingerprint Verification, Training, Artificial Intelligence, Simulink, Artificial Neural Network (ANN).