Palmprint identification using restricted fusion
Wei Jia
a,b
, Bin Ling
c,
*
, Kwok-Wing Chau
d
, Laurent Heutte
e
a
Intelligent Computation Lab, Hefei Institute of Intelligent Machines, Chinese Academy of Science, P.O. Box 1130, Hefei, Anhui 230031, China
b
Department of Automation, University of Science and Technology of China, Hefei 230027, China
c
Department of Obstetrics and Gynecology, Anhui Provincial Hospital, Anhui Medical University, Hefei 230001, China
d
Department of Civil and Structural Engineering, Hong Kong Polytechnic University, Hong Kong, China
e
Lab. LITIS, UFR Sciences, University of Rouen, France
article info
Keywords:
Palmprint identification
Biometrics
Fusion
Principal lines
Locality Preserving Projections
abstract
In this paper, we propose two palmprint identification schemes using fusion strategy. In
the first fusion scheme, firstly, the principal lines of test image is extracted, and matched
with that of all training images. Secondly, those training images with large matching scores
are selected to construct a small training sub-database. At last, the decision level fusion,
combing matching scores of principal lines and Locality Preserving Projections features,
has been made for final identification in small training sub-database. From another point
of view, it can be seen that the fusion is restricted by the previous results of principal lines
matching, so we call it as restricted fusion. The second fusion scheme is similar to the first
one. Just the fusion order is changed. The results of experiments conducted on PolyU palm-
print database show that the proposed schemes can achieve 100% accurate recognition
rate.
Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction
In information and networked society, automatic personal identification is an impending and crucial problem that needs
to be solved properly. As an efficient and safe solution, biometrics technology has recently been receiving wide attention
from researchers. Similar to fingerprint or iris based personal identification, palmprint based identification system can also
achieve good performance. For example, it can obtain high accurate recognition rates, and fast processing speed, etc. [1].At
the same time, palmprint based identification system has several special advantages such as stable line features, rich texture
features, low-resolution imaging, low-cost capturing devices, easy self-positioning, and user-friendly interface, etc. For these
reasons, nowadays the research related to this issue is becoming more active.
So far, there have been many approaches proposed for palmprint recognition including verification and identification,
which can be divided into five categories:
(1) Texture-based approaches have been studied extensively, which have shown good performance in terms of recogni-
tion rates and processing speed. Zhang et al. proposed PalmCode method, which employed one 2D Gabor filter to extract
the texture feature of palmprint [1]. Later, Kong et al. proposed FusionCode using feature-level fusion strategy, which can
be regarded as the improved version of the PalmCode [2]. Chen and Xie used dual-tree complex wavelets to extract tex-
ture energies of palmprint, and adopted SVM for classification [3]. Li and You proposed a texture-based palmprint retrie-
val scheme using a layered search strategy for personal identification [4].
0096-3003/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.amc.2008.05.024
* Corresponding author.
E-mail address: Bin.ling88@gmail.com (B. Ling).
Applied Mathematics and Computation 205 (2008) 927–934
Contents lists available at ScienceDirect
Applied Mathematics and Computation
journal homepage: www.elsevier.com/locate/amc