Design of a Reliable Quantization Scheme for Biometric Template Security

摘要：

Biometrics offers an alternative and reliable personal authentication system over conventional authentication methods as it is very unlikely for biometric sample to be lost or forgotten. However, the traditional methods of biometrics storage pose a number of security and privacy risk to users. Unlike conventional biometric systems which store the faithful representation of the captured biometric sample, the proposed statistical quantization transformation (SQT) method transforms the biometric sample into bit-string of template and stores a copy of the transformed template. The proposed method limits the potential invasion of biometric template as authentication takes place in the transformed domain rather than the original feature space. The proposed SQT is evaluated using fingerprint dataset and our experimental and security analysis suggests its feasible implementation in practice.

关键词： Statistical Quantization Transformation Biometrics authentication privacy fingerprint

作者: Thian Song Ong Andrew Beng Jin Teoh Fei Tan Connie Tee

作者单位: Faculty of Information Science and Technology, Multimedia University, Malaysia School of Electrical and Electronic Engineering, Yonsei University, Seoul, Korea

会议类型: 国际会议

会议名称: 2010 International Conference on Information Security and Artificial Intelligence(2010年信息安全与人工智能国际会议 ISAI 2010)

会议地点: 成都

会议语种:英文

页码: 1319-1324

在线出版日期: 2010-12-17（万方平台首次上网日期，不代表论文的发表时间）