Discovering Useful Information for Personal Authentication by Statistical Analyses on Biometric Information

本研究试图通过对生物特征信息的统计分析，为个人身份认证寻找有用的信息，从而实现一个安全、实用的基于生物特征的身份认证系统。本研究以掌纹图像和指纹图像为研究对象，通过对样本图像间相似度的统计分析，提出了一种准确高效的身份认证系统，能够快速检索到查询图像的相似图像。根据这些结果，可以估计出生物特征图像中用于个人身份认证的有用信息的维度，这有望指导我们发现这些信息。

This research is trying to find useful information for personal authentication by statistical analyses on biometric information in order to realize a secure and useful biometric-based authentication system. This research focused on palmprint images and fingerprint images, and proposed an accurate and efficient authentication system which can search a similar image of a query image fast by a statistical analysis on the similarities between sample images. By the results, the dimension of useful information for personal authentication in biometric images can be estimated, which is expected to lead we are able to discover the information.