基于口令的加密模糊金库  

Encrypted Fuzzy Vault Based on Password

作　　者：秦彩霞[1] 游林[1] 

机构地区：[1]杭州电子科技大学通信工程学院,杭州310018

出　　处：《科技通报》2011年第2期282-287,共6页Bulletin of Science and Technology

基　　金：国家自然科学基金项目(60763009)

摘　　要：模糊金库已成为生物特征识别技术与密码技术相结合的一个重要方案。其安全性仅是基于在大量无序的特征点中选取正确的细节特征点并利用它们来重建多项式的难度。为了进一步提高系统安全性,提出加密模糊金库方案。先利用口令加密指纹细节特征点,再将加密后的细节特征点应用于模糊金库的加锁和解锁过程中。攻击者即使在大量无序的特征点中选取出了正确的细节特征点并利用它们重建了多项式,没有输入正确口令时,也得不到密钥。在FVC2002-DB2指纹数据库上与原始模糊金库进行了仿真比较。实验结果表明,加密模糊金库的错误接受率(FAR)从0.5%下降为0.001%,正确接收率(GAR)从95%上升到97%。Fuzzy vault has become an important scheme combining biometric technology with cryptography. Its security depends only on the difficulty of identifying the genuine minutiae points from a mass of disordered points and reconstructing a polynomial with the genuine minutiae points. Aiming at higher security of the system,encrypted fuzzy vault scheme was proposed, which encrypted the minutiae points first and then encoded and decoded fuzzy vault using the encrypted minutiae points. Even though an attacker can identify the genuine minutiae points from the disordered points and reconstruct the polynomial with them, he cannot get the key without the correct password. Experimental comparisons with primitive fuzzy vault on FVC2002-DB2 fingerprint database show that the false accept rate （FAR） of encrypted fuzzy vault is reduced from 0.5% to 0.001% ,and the genuine accept rate（GAR） is increased from 92.5% to 93.5%.

关 键 词：口令 模糊金库 生物特征识别 指纹细节点 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]