[1]LI Qian-ying,RUAN Qiu-qi.Palmprint feature extraction based on multiresolution LBP[J].CAAI Transactions on Intelligent Systems,2010,5(6):482-486.
Copy
CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume:
5
Number of periods:
2010 6
Page number:
482-486
Column:
学术论文—人工智能基础
Public date:
2010-12-25
- Title:
-
Palmprint feature extraction based on multiresolution LBP
- Author(s):
-
LI Qian-ying; RUAN Qiu-qi
-
Department of Computer Science and Technology, Northern Jiaotong University, Beijing 100044, China
-
- Keywords:
-
palmprint identification; NSCT; LBP; shiftinvariant; rotation invariant
- CLC:
-
TP391.41
- DOI:
-
-
- Abstract:
-
In order to study the feasibility and performance of the nonsubsampled contourlet(NSCT) and local binary pattern(LBP)operation in palm recognition, this paper presented a palmprint identification approach, which was the combinations of NSCT and LBP operation. It can effectively extract detail features of palmprint and reduce the influence on the shift, rotation and illumination of palmprint identification. The NSCT provides sparse representation at both spatial and directional resolution and is a fully shiftinvariant. LBP is a powerful texture descriptor that is grayscale and rotation invariant.In this paper,we extracted multiresolution and directional local binary pattern histogram to represent the local features of a palmprint image after NSCT. The results of experiments illustrate that this identification approach is efficient and effectively extract details and structure of palmprint.