摘要
针对基于单一模态的手指生物特征身份认证技术在实际应用中受数据噪声、类内差异、类间相似及防伪性低等因素制约的问题,提出基于判别相关分析的手指多模态融合身份识别方法。以手指静脉为研究对象,结合指腹纹理特征,采用广义判别分析法降维,并采用判别相关分析法对降维后的指静脉和指腹纹理特征进行融合,实现了基于特征层融合的手指多模态生物特征的身份认证识别。本方法实现了手指静脉和指腹纹特征的融合,与传统的核线性分析和典型相关分析方法相比,提高了身份认证的准确率,促进了手指多模态生物特征认证的进一步发展。
The finger-based biometrics authentication technology based on single mode is inevitably limited by data noise, in- tra-class difference, similarity between classes and low security, a multi-modal fusion recognition method based on discrimi- nant correlation analysis is proposed. With finger vein as the research object, combined with the features of finger texture, the generalized discriminant analysis method is used to reduce the dimension, and discriminant correlation analysis is adopted for fusing two modal features after dimension reduction, and the identity authentication of finger multi-modal biometrics based on feature layer fusion is realized. This method realizes the fusion of finger vein and finger dorsal texture, which im- proves the accuracy of identity authentication and the further development of finger multi modal biometric authentication compared with the traditional nuclear linear analysis and typical correlation analysis method.
作者
胡锦丽
杨辉华
刘振丙
HU Jinli;YANG Huihua;LIU Zhenbing(School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China;School of Computer and Information Security, Guilin University of Electronic Technology, Guilin 541004, China;School of Automation, Beijing University of Posts and Telecommunications, Beijing 100876, China)
出处
《桂林电子科技大学学报》
2017年第6期431-436,共6页
Journal of Guilin University of Electronic Technology
基金
国家自然科学基金(21365008
61562013)
广西重点研发计划项目(桂科AB16380293)
关键词
指静脉
多模态
降维
判别相关分析
finger vein
multi-modal
dimensionality reduction
diseriminant correlation analysis
