Finger Vein Recognition with Gabor Wavelets and Local Binary Patterns

Jialiang PENG; Qiong LI; Ahmed A. ABD EL-LATIF; Ning WANG; Xiamu NIU

doi:10.1587/transinf.E96.D.1886

Finger Vein Recognition with Gabor Wavelets and Local Binary Patterns

Jialiang PENG, Qiong LI, Ahmed A. ABD EL-LATIF, Ning WANG, Xiamu NIU

Full Text Views

0

Share
Cite this

Summary :

In this paper, a new finger vein recognition method based on Gabor wavelet and Local Binary Pattern (GLBP) is proposed. In the new scheme, Gabor wavelet magnitude and Local Binary Pattern operator are combined, so the new feature vector has excellent stability. We introduce Block-based Linear Discriminant Analysis (BLDA) to reduce the dimensionality of the GLBP feature vector and enhance its discriminability at the same time. The results of an experiment show that the proposed approach has excellent performance compared to other competitive approaches in current literatures.

Publication: IEICE TRANSACTIONS on Information Vol.E96-D No.8 pp.1886-1889

Publication Date: 2013/08/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E96.D.1886

Type of Manuscript: LETTER

Category: Pattern Recognition

Authors

Jialiang PENG
  Harbin Institute of Technology,Heilongjiang University
Qiong LI
  Harbin Institute of Technology
Ahmed A. ABD EL-LATIF
  Harbin Institute of Technology,Menoufia University
Ning WANG
  Harbin Institute of Technology
Xiamu NIU
  Harbin Institute of Technology

Keyword

finger vein recognition, Gabor wavelet, local binary pattern, BLDA

Cite this

Copy

Jialiang PENG, Qiong LI, Ahmed A. ABD EL-LATIF, Ning WANG, Xiamu NIU, "Finger Vein Recognition with Gabor Wavelets and Local Binary Patterns" in IEICE TRANSACTIONS on Information, vol. E96-D, no. 8, pp. 1886-1889, August 2013, doi: 10.1587/transinf.E96.D.1886.
Abstract: In this paper, a new finger vein recognition method based on Gabor wavelet and Local Binary Pattern (GLBP) is proposed. In the new scheme, Gabor wavelet magnitude and Local Binary Pattern operator are combined, so the new feature vector has excellent stability. We introduce Block-based Linear Discriminant Analysis (BLDA) to reduce the dimensionality of the GLBP feature vector and enhance its discriminability at the same time. The results of an experiment show that the proposed approach has excellent performance compared to other competitive approaches in current literatures.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E96.D.1886/_p

Copy

@ARTICLE{e96-d_8_1886,
author={Jialiang PENG, Qiong LI, Ahmed A. ABD EL-LATIF, Ning WANG, Xiamu NIU, },
journal={IEICE TRANSACTIONS on Information},
title={Finger Vein Recognition with Gabor Wavelets and Local Binary Patterns},
year={2013},
volume={E96-D},
number={8},
pages={1886-1889},
abstract={In this paper, a new finger vein recognition method based on Gabor wavelet and Local Binary Pattern (GLBP) is proposed. In the new scheme, Gabor wavelet magnitude and Local Binary Pattern operator are combined, so the new feature vector has excellent stability. We introduce Block-based Linear Discriminant Analysis (BLDA) to reduce the dimensionality of the GLBP feature vector and enhance its discriminability at the same time. The results of an experiment show that the proposed approach has excellent performance compared to other competitive approaches in current literatures.},
keywords={},
doi={10.1587/transinf.E96.D.1886},
ISSN={1745-1361},
month={August},}

Copy

TY - JOUR
TI - Finger Vein Recognition with Gabor Wavelets and Local Binary Patterns
T2 - IEICE TRANSACTIONS on Information
SP - 1886
EP - 1889
AU - Jialiang PENG
AU - Qiong LI
AU - Ahmed A. ABD EL-LATIF
AU - Ning WANG
AU - Xiamu NIU
PY - 2013
DO - 10.1587/transinf.E96.D.1886
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2013
AB - In this paper, a new finger vein recognition method based on Gabor wavelet and Local Binary Pattern (GLBP) is proposed. In the new scheme, Gabor wavelet magnitude and Local Binary Pattern operator are combined, so the new feature vector has excellent stability. We introduce Block-based Linear Discriminant Analysis (BLDA) to reduce the dimensionality of the GLBP feature vector and enhance its discriminability at the same time. The results of an experiment show that the proposed approach has excellent performance compared to other competitive approaches in current literatures.
ER -