IEICE Transactions on Information and Systems
Online ISSN : 1745-1361
Print ISSN : 0916-8532
Regular Section
Feature-Level Fusion of Finger Veins and Finger Dorsal Texture for Personal Authentication Based on Orientation Selection
Wenming YANGGuoli MAFei ZHOUQingmin LIAO
著者情報
ジャーナル フリー

2014 年 E97.D 巻 5 号 p. 1371-1373

詳細
抄録
This study proposes a feature-level fusion method that uses finger veins (FVs) and finger dorsal texture (FDT) for personal authentication based on orientation selection (OS). The orientation codes obtained by the filters correspond to different parts of an image (foreground or background) and thus different orientations offer different levels of discrimination performance. We have conducted an orientation component analysis on both FVs and FDT. Based on the analysis, an OS scheme is devised which combines the discriminative orientation features of both modalities. Our experiments demonstrate the effectiveness of the proposed method.
著者関連情報
© 2014 The Institute of Electronics, Information and Communication Engineers
前の記事 次の記事
feedback
Top