TY - GEN
T1 - Cancellable fingerprint generation by projection based transformation and quantization
AU - Gedalia Razafindrobelina, Nabe
AU - Ahmad, Tohari
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/9/19
Y1 - 2020/9/19
N2 - In daily life, biometrics has been more common to use for authentication purposes. Nevertheless, the security of the biometric data and the corresponding process still challenge that the public must not access them. Thus, the demand for secure, revocable, and cancellable templates is increasing. The biometric protection templates can be done in general by two methods: Bio-cryptosystems and cancellable biometrics. Some existing methods suffer from performance problems, such as the complexity and the verification error. In this paper, a method to generate a cancellable template is proposed, which uses a projection transformed minutiae. The security is approached by mapping the data to two 3D arrays for generating two-bit strings then after a Fourier transformation is applied. The resulting two complex vectors are multiplied to form the cancellable template. Here, the security of the template is improved by the use of a user-specific or general key. The extracted features do not need to be aligned, because the minutiae by itself are the reference points and the neighbors inside a range projected into the projection line. Additionally, for the mapping of the projected coordinates, the 3D array is divided into three partitions. The middle partitions are with higher resolution, because of the probability of projected point positions. After that transformations, the template is non-invertible and revocable. The experiment shows that this method is the potential to extend for achieving the authentication purpose.
AB - In daily life, biometrics has been more common to use for authentication purposes. Nevertheless, the security of the biometric data and the corresponding process still challenge that the public must not access them. Thus, the demand for secure, revocable, and cancellable templates is increasing. The biometric protection templates can be done in general by two methods: Bio-cryptosystems and cancellable biometrics. Some existing methods suffer from performance problems, such as the complexity and the verification error. In this paper, a method to generate a cancellable template is proposed, which uses a projection transformed minutiae. The security is approached by mapping the data to two 3D arrays for generating two-bit strings then after a Fourier transformation is applied. The resulting two complex vectors are multiplied to form the cancellable template. Here, the security of the template is improved by the use of a user-specific or general key. The extracted features do not need to be aligned, because the minutiae by itself are the reference points and the neighbors inside a range projected into the projection line. Additionally, for the mapping of the projected coordinates, the 3D array is divided into three partitions. The middle partitions are with higher resolution, because of the probability of projected point positions. After that transformations, the template is non-invertible and revocable. The experiment shows that this method is the potential to extend for achieving the authentication purpose.
KW - authentication
KW - biometrics
KW - cancellable template
KW - fingerprint
KW - projection-based transformation
UR - http://www.scopus.com/inward/record.url?scp=85096842253&partnerID=8YFLogxK
U2 - 10.1109/iSemantic50169.2020.9234192
DO - 10.1109/iSemantic50169.2020.9234192
M3 - Conference contribution
AN - SCOPUS:85096842253
T3 - Proceedings - 2020 International Seminar on Application for Technology of Information and Communication: IT Challenges for Sustainability, Scalability, and Security in the Age of Digital Disruption, iSemantic 2020
SP - 286
EP - 291
BT - Proceedings - 2020 International Seminar on Application for Technology of Information and Communication
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 International Seminar on Application for Technology of Information and Communication, iSemantic 2020
Y2 - 19 September 2020 through 20 September 2020
ER -