摘要
小样本问题是人脸识别系统中的一个常见问题,多模态模型具有很强的泛化能力去解决小样本问题,并且已经成为模式识别中重要的研究领域,但是该模型的低精度及低效率已经成为目前的主要难题。为了解决这个问题,提出了一个高效的基于神经网络的多模生物特征人脸及指纹识别系统,通过选择好的特征提取及识别算法来提供更高效的识别。采用生物特征识别来验证一个人的身份,相对于采用密码或口令大大提高了系统的运行效率和识别精度。
The problem of the small sample size can be met frequently in face recognition system, multi-model model has strong generalization ability to solve the problem of the small sample size and has become the most important field research in pattern recognition. However, both precision and low efficiency is the main challenge currently. To address this problem, an efficient muhimode biometric face and fingerprint identification system is proposed by selecting a good feature extraction and recognition algorithm to provide more precise identification. Using biometrics to verify the identity of a person greatly improving the efficiency of the system and the recognition accuracy compared to the use of a password or password.
出处
《电视技术》
北大核心
2014年第5期178-180,189,共4页
Video Engineering
基金
河南省基础与前沿技术研究项目(112300410225)
关键词
人脸识别
特征抽取
多层次感应
指纹识别
神经网络
face recognition
feature extraction
muhilayer perception
fingerprint recognition
neural network
