TY - JOUR
T1 - Combining touched fingerprint and finger-vein of a finger, and its usability evaluation
AU - Nguyen, Dat Tien
AU - Park, Young Ho
AU - Lee, Hyeon Chang
AU - Shin, Kwang Yong
AU - Kang, Byung Jun
AU - Park, Kang Ryoung
PY - 2012/1
Y1 - 2012/1
N2 - Although fingerprint and finger-vein recognition systems are widely used in various applications, their performance is limited by their sensitivity to wet or dry fingers, skin distortions, scars on the fingers, and changes in venous blood flow. To overcome these problems, we propose a new multimodal biometric system based on touched fingerprint and finger-vein recognition. This research is novel in the following four ways. First, the device proposed herein acquires fingerprint and finger-vein images simultaneously. The fingerprint image is obtained from the region between the fingertip and the first finger joint, whereas the finger-vein image is acquired from the area between the first and the second finger joint. The proposed device can easily be implemented by attaching a commercial fingerprint sensor to a laboratory-made finger-vein sensor, which eliminates the need for complicated configurations of these two sensors. Second, the misalignments of fingerprint and finger-vein images can be reduced through touching the finger area between the fingertip and second knuckle on the surface of device. Third, by combining the distances of fingerprint and finger-vein recognition based on a support vector machine (SVM), the consequent combined accuracy is markedly enhanced. Fourth, through a subjective test, the usability of the proposed device is compared with that of conventional fingerprint and finger-vein sensors. The experimental results show that the equal error rate (EER) of the proposed method was 0.5%, which is lower than that of fingerprint or finger-vein recognition alone. Moreover, the usability of the proposed device is comparable to that of the fingerprint and finger-vein sensors.
AB - Although fingerprint and finger-vein recognition systems are widely used in various applications, their performance is limited by their sensitivity to wet or dry fingers, skin distortions, scars on the fingers, and changes in venous blood flow. To overcome these problems, we propose a new multimodal biometric system based on touched fingerprint and finger-vein recognition. This research is novel in the following four ways. First, the device proposed herein acquires fingerprint and finger-vein images simultaneously. The fingerprint image is obtained from the region between the fingertip and the first finger joint, whereas the finger-vein image is acquired from the area between the first and the second finger joint. The proposed device can easily be implemented by attaching a commercial fingerprint sensor to a laboratory-made finger-vein sensor, which eliminates the need for complicated configurations of these two sensors. Second, the misalignments of fingerprint and finger-vein images can be reduced through touching the finger area between the fingertip and second knuckle on the surface of device. Third, by combining the distances of fingerprint and finger-vein recognition based on a support vector machine (SVM), the consequent combined accuracy is markedly enhanced. Fourth, through a subjective test, the usability of the proposed device is compared with that of conventional fingerprint and finger-vein sensors. The experimental results show that the equal error rate (EER) of the proposed method was 0.5%, which is lower than that of fingerprint or finger-vein recognition alone. Moreover, the usability of the proposed device is comparable to that of the fingerprint and finger-vein sensors.
KW - Finger-vein recognition
KW - Fingerprint recognition
KW - Misalignment
KW - Multimodal biometric
KW - Usability
UR - http://www.scopus.com/inward/record.url?scp=84861524877&partnerID=8YFLogxK
U2 - 10.1166/asl.2012.2177
DO - 10.1166/asl.2012.2177
M3 - Article
AN - SCOPUS:84861524877
SN - 1936-6612
VL - 5
SP - 85
EP - 95
JO - Advanced Science Letters
JF - Advanced Science Letters
IS - 1
ER -