A Fingerprint and Voiceprint Fusion Identity Authentication Method

摘要

In the traditional biometric identification scheme, the single fingerprint feature points are used for identification, or the single voiceprint is used as the authentication standard, and it is difficult to obtain a good accuracy in a complicated environment. Different biometrics have different advantages, disadvantages and applicable scenarios. A single mode cannot have a wider coverage scene. For this reason, we propose a fusion algorithm for fingerprint recognition and voiceprint recognition, combining the recognition characteristics of fingerprint and voiceprint, an identification scheme based on fingerprint and voiceprint fusion is proposed. The eigenvalues of fingerprint and voiceprint are divided into a group. The depth neural network is used to extract the fingerprint and voiceprint features respectively, and the probability combination is used to verify the fusion. The experimental results show that the combination of the two certifications reduces the error acceptance rate (FAR) by 4.04% and the error rejection rate (FRR) by 1.54% compared to a single fingerprint or voice-print recognition scheme.