Research On Dual Mode Identity Authentication Based On Fingerprint And ECG

Fingerprint based biometric technology is widely used in various fields of society,but the incomplete fingerprint will lead to the decline of accuracy,and the fingerprint is easy to be forged,the system is vulnerable to deception attacks,there are security risks.The ECG signal has the characteristics of living body,which is not easy to be copied and forged,and has higher security.Therefore,in this paper,fingerprint and ECG are fused to improve the security of the identity authentication system.To solve the problem of accuracy degradation caused by incomplete fingerprint and deception attack caused by false fingerprint film,a dual-mode fusion framework based on D-S theory is proposed in this paper,which fuses fingerprint and ECG signal in fractional layer.The experimental results show that the false rejection rate of the traditional feature point matching algorithm is 100% when the incomplete fingerprint feature information is insufficient,while the proposed dual-mode fusion authentication reduces the false rejection rate to 5%,which improves the robustness of the system;the framework can effectively solve the evidence conflict problem by using D-S decision rules,when the false fingerprint film passes the authentication but the ECG signal fails,Finally,the system gives the result of correct rejection.The validation was performed on the data set constructed by ECG-ID and FVC2002 database.The error acceptance rate of 22 groups of data is less than 7%,which greatly improves the security of the system.To ensure that the collected ECG signal comes from the living body,this paper proposes a steganography based ECG detection module based on the fusion framework.The module uses LSB steganography algorithm to embed the ECG dynamic feature data extracted from the acquisition terminal into the original ECG signal.Because the embedded dynamic feature data is different each time,it is difficult for attackers to imitate.It is verified on ECG-ID data set.The experimental results show that this module can ensure the accuracy of identity authentication and improve the security of system authentication.