Design and implementation of secure fingerprint template using key binding scheme and random triangle hashing scheme

One problem in biometric fingerprint technology is template protection. Template is an individual's digital reference extracted from biometric sample. The objective of template protection is to protect user biometric information. Basis idea of template protection is if an attacker able to get the template then the attacker will not be able to reverse template to obtain biometric information. Template protection also offer revocability for users. This research use modified Key Binding scheme and Random Triangle scheme to protect biometric template. The main advantage of our proposed design is alignment free. Performance of our Key Binding scheme is still very poor (EER = 30%). Nearest Neighbors Structure used for local matching in our Key Binding scheme still cannot replace alignment process. ECC and Hash module give a significant contribution to higher EER. Random Triangle Hashing (RTH) is successfully implemented with EER = 0.96%. This performance is better than the winner of FVC2004 (EER = 1.97%). We modified RTH in the transformation module (NNS) and evaluation score (Hamming distance).