TY  - JOUR
T1  - An Efficient Technique for Human Verification Using Finger Stripes Geometry
AU - , Arafatur Rahman AU - , Saiful Azad AU - , Farhat Anwar 
JO  - International Journal of Soft Computing
VL  - 2
IS  - 3
SP  - 445
EP  - 449
PY  - 2007
DA  - 2001/08/19
SN  - 1816-9503
DO  - ijscomp.2007.445.449
UR  - https://makhillpublications.co/view-article.php?doi=ijscomp.2007.445.449
KW  - Biometrics
KW  -feature extraction
KW  -back propagation algorithm
KW  -distance based nearest neighbors algorithms
KW  -human verification
KW  -artificial neural network
AB  - This study attempts to improve the performance of person verification by finger stripes geometry, which is an efficient, simple, fast, easy to handle and cost effective compared with other biometric human verification technique. Finger stripe geometry is a new biometric type in verification process. This finger stripe based verification consists of two main attributes, feature extraction by image processing and feature learning by ANN (Artificial Neural Network). The Distance Based Nearest Neighbor Algorithm, which shows greater accuracy than NN is also applied here. Using this approach, experimental result provides, Total Success Rate (TSR) is 98.79%, False Acceptance Rate (FAR) is 3.21% and False Rejection Rate (FRR) is 0%. Since the proposed method requires very small computational time, the practical implementation requirement for a large number of person verification is possible.
ER  -