【作者】 袁立； 张金辉； 穆志纯； 杨帆；

【Author】 YUAN Li~1,ZHANG Jin-hui~1,MU Zhi-chun~1,YANG Fan~2 (1.School of Automation and Electrical Engineering, University of Science and Technology Beijing,Beijing 100083,China; 2.LE2I Laboratory,University of Burgundy,21078 Dijon Cedex,France)

【机构】 北京科技大学 自动化学院； LE2I Laboratory,University of Burgundy,21078 Dijon Cedex；

【摘要】 以人耳作为生物特征,将模糊保险箱(Fuzzy vault)算法应用到人耳生物特征模板的保护中。首先,利用基于代数特征的主元分析法提取人耳图像的特征向量;然后,利用模糊保险箱算法实现模板保护,该过程分为加密和解密两部分。在加密过程中,对密钥进行循环冗余加密,加密后构造多项式函数;然后,将人耳特征数据在该多项式上投影,并添加一定数量的杂凑点,构成保险箱;在解密过程中,利用保险箱和待认证人耳的生物特征模板构建候选点集合,利用有限域内的拉格朗日插值法解出多项式,并用循环冗余编码对多项式系数进行校验,若满足校验结果则恢复密钥成功。在北京科技大学人耳图像库(USTB人耳图像库)上的实验结果表明,该方法能够有效进行模板加密。更多还原

【Abstract】 An protection method was proposed for ear biometric template based on fuzzy vault.Feature templates of the training samples were firstly extracted using PCA.In order to secure an ear biometric template,an uniformly random cryptographic key was generated and this key was transformed into a polynomial.All the elements in the ear template were then evaluated on this polynomial.This set of points was then secured by hiding them among a large set of q randomly generated chaff points that do not lie on the polynomial.The set of genuine and chaff points along with their polynomial evaluations constituted the vault.During authentication,if the query biometric set was sufficiently close to the training ear template,the polynomial could be successfully reconstructed by identifying the genuine points in the vault that were associated with the training ear template.Experimental results on the USTB ear database show that fuzzy vault can be successfully applied to ear template encryption.更多还原