CORC  > 自动化研究所  > 中国科学院自动化研究所  > 毕业生  > 博士学位论文
题名基于指纹认证的加密技术研究与应用
作者李亮
学位类别工学博士
答辩日期2007-06-18
授予单位中国科学院研究生院
授予地点中国科学院自动化研究所
导师田捷
关键词指纹识别 指纹加密 纠错编码 身份认证 安全邮件 Fingerprint recognition Fingerprint encryption Error-correcting code Identity authentication Secure Email
其他题名The Research and Application on Fingerprint Authentication Based Encryption
学位专业计算机应用技术
中文摘要随着我国信息化建设的不断深入,互联网的普遍应用所带来的信息安全问题也越来越引起人们的重视。信息安全技术的核心是现代密码理论的研究与应用。然而无论是PKI还是IBE,所针对的都是用户的数字身份,尽管加密密钥能够确保信息安全系统中数字身份的安全,但由于这些密钥通常是很长且难以记忆的随机数据,因而需要将密钥保存在诸如智能卡或USB钥匙等存储介质中,再以某种替代的认证方式通过对存储介质的访问来获得密钥。这些替代的认证方式通常就是密码或者口令,因此整个系统的安全性和身份认证其实是基于密码或口令的,密码或口令的低安全性导致了现有的技术无法建立起数字身份与合法用户之间的有效联系。 本文主要研究一种基于指纹的加密方法,通过指纹认证来实现密钥的合法访问和管理,并针对网络环境下的应用场合提出一种基于指纹加密的网络环境身份认证方案,最后基于这一方案研究实现了一种指纹安全邮件系统。 本文的主要研究工作概括如下: 1)提出了一种基于指纹频域的纠错指纹加密算法。为了解决加密算法中的指纹配准问题,首先提出了一种基于脊线跟踪的参考点定位方法用于指纹配准;其次选择小波分解与傅立叶-梅林变换提取指纹的频域特征,称之为WFMT(Wavelet Fourier-Mellin Transformation)特征,并使用这种特征进行向量量化编码,编码后的指纹特征可以在汉明距测度空间下度量;最后设计了基于纠错编码的纠错指纹加密算法的加/解密过程。实验结果表明本文提出的方法能够可靠保护指纹特征与加密密钥,并实现了基于指纹认证的密钥管理,且准确率优于目前文献中的基于指纹加密的方法。 2)提出了一种基于指纹加密的网络环境身份认证方案,称之为FIBE(Fingerprint & Identity Based Encryption)。该方案结合了IBE的加密算法和签名算法并提出一种指纹摘要匹配算法应用于网络环境下的身份认证。通过指纹摘要匹配算法,信息接收方可以在不需可信第三方的参与下对信息发送方进行远程指纹匹配从而认证信息发送方的身份,因此该方案适合运用于电子政务环境中的垂直认证体系。同时该方案还设计了用户通过指纹认证向可信第三方在线获取私钥和在线更新私钥,流程简易方便。该方案达到了秘密通信系统对机密性、完整性、不可否认性、可控性等要求,可作为PKI等其它网络认证方案的补充。 3)提出了一种基于指纹认证的安全邮件系统,该系统构建了一个基于指纹认证的安全邮件通信环境,实现了可靠的身份认证和数据传输,是所提出的基于指纹加密的网络环境身份认证方案的一个应用实例。同时为了处理现今网络应用中可能存在的多种指纹采集设备共同使用的问题,首次提出了基于多类型指纹传感器的交叉比对算法,通过图像空间归一化和特征空间归一化两级变换方法,将不同传感器采集的图像的特征点映射到同一个特征空间下,最后在这个特征空间下进行特征点匹配。实验结果表明这种方法能够归一化处理目前具有代表性的指纹采集设备(光学传感器、电感传感器和热敏传感器)的指纹图像,并具有较高的识别性能。
英文摘要With the rapid progress of informationization, security problems caused by networking transactions have raised more concerns. Currently the certificate-based information security techniques face the problem of comprehensive management of certificates and high computation burden, and fret the high cost of implementation and maintenance. Identity Based Encryption (IBE) is a more flexible encryption scheme, which can not only simplify the problems such as user registration, management and revoke of certificates, but provide the equal security and secrecy to PKI. However, either PKI or IBE just operate on digital identities, although cryptographic keys can ensure their security in cryptosystem, they are often accessed by alternative authentication such as password. Password-based authentication results in the weak connection between digital identity and legitimate user. This paper studies the seamless integration of fingerprint recognition with cryptosystem, and tries to build a stronger connection between cryptographic key with users. The main work of this dissertation is as follows: 1) We propose an error-correcting fingerprint encryption algorithm based on fingerprint frequency spectrum. First, a reference point detection method based on ridge-tracking is proposed to align two fingerprints. Second, we deploy Fourier-Mellin transformation to extract frequency features of fingerprint image. Then the frequency features are coded into binary bit sequence by Vector Quantization. Finally we construct a key binding algorithm based on error-correcting technique to retrieve the key by successful fingerprint authentications. The proposed method can also protect fingerprints and keys safely and experimental result shows its good performance. 2) We propose a fingerprint based networking identity authentication scheme. This scheme proposes a fingerprint hash matching algorithm and combines the encryption and signature algorithm of IBE. It constructs a secure communication environment, in which legitimate users can authenticate the identity of users by fingerprint hash without a trustworthy third-party participating online. It is suitable to be applied in occasions of vertical identity authentication in e-government transactions and can be a complementarity of PKI. 3) We proposed a secure fingerprint-based Email system. It is an instance of our proposed networking identity authentication scheme. The secure communication environment is constructed and assured by secure identity authentication and secure data encryption. We also proposed a novel cross-matching algorithm to handle the specific characteristics of fingerprint images from different type of sensors. By a two-level transformation–image space normalization and feature space normalization, all feature points of fingerprint images are mapped into one feature space. Experimental results show that this method achieves good performance on image datasets of optical,thermal and capacity sensors.
语种中文
其他标识符200418014628058
内容类型学位论文
源URL[http://ir.ia.ac.cn/handle/173211/6025]  
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
李亮. 基于指纹认证的加密技术研究与应用[D]. 中国科学院自动化研究所. 中国科学院研究生院. 2007.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace