【摘要】 随着人脸识别在门禁、视频监控等公共安全领域中的应用日益广泛,人脸特征数据的安全性和隐私性问题成为备受关注的焦点。近年来出现了许多关于生物特征及人脸特征的安全保护算法,这些算法大都是将生物特征数据转变为二值的串,再进行保护。针对已有的保护算法中将实值的人脸特征转换为二值的串,从而导致信息丢失的不足,应用模糊逻辑对人脸模板数据的类内差异进行建模,从而提高人脸识别系统的性能。给出了算法在CMU PIE的光照子集、CMU PIE带光照和姿势的子集和ORL人脸数据库中的实验结果。实验表明,该算法能够进一步提高已有安全保护算法的识别率。更多还原

【Abstract】 With the growing use of face recognition in security and video control domain,there is growing concern about the security and privacy of the biometrics data.Recently,technologies for biometric security and privacy have been proposed,which typically transform the biometric data to a binary string.These transformations can lead to some information loss and downgrade the performance of a system.This paper applied fuzzy logic to confirm the reliability of each bit in a binary string and to model the intra-class variations.The experimental results show this method reduces the overlap of imposter distribution and genuine distribution and improves the performance of biometric security technology.更多还原