Cryptographic Authentication from the Iris
Cryptographic Authentication from the Iris
复制标题
DOI：
10.1007/978-3-030-30215-3_23
复制
发表时间：
2019-09
期刊：
影响因子：
--
通讯作者：
Sailesh Simhadri;James Steel;Benjamin Fuller
中科院分区：
文献类型：
--
作者：
Sailesh Simhadri;James Steel;Benjamin Fuller
生物测定显示重复读数之间存在噪声。由于噪声，设备存储了生物识别的明文模板。该存储的模板是攻击者的诱人目标。模糊抽取器从生物测定中获得稳定的密钥(Dodis等人，Eurocrypt 2004)。尽管进行了多次尝试，但还没有证明密钥强度下限的虹膜密钥推导系统。我们的起点是由于Canetti等人的模糊提取(Eurocrypt 2016)。我们对图像处理和密码算法进行了修改和耦合。给出了虹膜分布安全的一个充分条件，并利用ND0405虹膜数据集对此条件进行了分析，构建了一个即使多个密钥来自同一虹膜也具有一定安全性的虹膜密钥派生系统。我们承认，对于一个安全的系统来说，一点安全是不够的。多因素系统为密码认证带来了最大的希望。我们的方案适合于加入额外的无噪声因素，如密码。我们的方案是用C和Python语言实现的，并且是开源的。
Biometrics exhibit noise between repeated readings. Due to the noise, devices store a plaintexttemplateof the biometric. This stored template is an appetizing target for an attacker.Fuzzy extractors derive a stable cryptographic key from biometrics (Dodis et al., Eurocrypt 2004). Despite many attempts, there are no iris key derivation systems that prove lower bounds on key strength.Our starting point is a fuzzy extractor due to Canetti et al. (Eurocrypt 2016). We modify and couple the image processing and cryptographic algorithms. We then present a sufficient condition on the iris distribution for security, and analysis this condition using the ND0405 Iris dataset.We build an iris key derivation system withbits of security even when multiple keys are derived from the same iris. We acknowledgebits of security is insufficient for a secure system. Multifactor systems hold the most promise for cryptographic authentication. Our scheme is suited for incorporation of additionalnoiselessfactors such as a password.Our scheme is implemented in C and Python and is open-sourced.