摘要： 传统的身份识别系统利用单一的生物特征作为依据，在复杂背景下，系统性能往往会大幅下降。基于数据融合的多生物特征身份识别技术可以提高生物识别系统的准确率等性能。该文利用特征脸和矢量量化方法建立人脸识别和语音识别两个子系统，在决策层用神经网络融合子系统的输出来进行身份识别。实验证明该方法比单个子系统识别率高，在噪音环境下，优势明显。

关键词: 多生物特征, 身份识别, 数据融合, 神经网络

Abstract: Traditional personal identification system based on single biometric (such as face, fingerprint, voice) is often not able to meet the requirements of system. The multi-biometric technology can improve the recognition accuracy comparing the single biometric feature system. This paper presents a multi-biometrics system, which integrates face and acoustic subsystem using neuron network. Experiments show that the fusion system performs well both in response time and system accuracy especially in noisy background.

Key words: Multi-biometric, Personal identification, Data fusion, Neuron network

中图分类号: 

• TP18