Palmprint verification using binary orientation co-occurrence vector
Palmprint verification using binary orientation co-occurrence vector
复制标题
DOI：
10.1016/j.patrec.2009.05.010
复制
发表时间：
2009-10-01
影响因子：
5.1
通讯作者：
Zuo, Wangmeng
中科院分区：
文献类型：
--
作者：
Guo, Zhenhua;Zhang, David;Zuo, Wangmeng
开发准确、健壮的掌纹识别算法是掌纹自动识别系统中的一个关键问题。在众多的掌纹识别方法中，基于方向的编码方法，如竞争性编码(CompCode)、掌纹方向编码(POC)和稳健行方向编码(RLOC)，是目前研究的热点。它们提取并编码局部优势方向作为特征，能够实时、高精度地匹配输入的掌纹。然而，只使用一个主方向来表示局部区域可能会丢失一些有价值的信息，因为掌纹中有交叉线。本文提出了一种新的特征提取算法，即二值方向共生向量(BOCV)，用于表示局部区域的多个方向。BOCV能够更好地描述图像的局部方向特征，对图像旋转具有更强的鲁棒性。在公共掌纹数据库上的实验结果表明，该算法在显著降低等错误率(EER)方面优于CompCode、POC和RLOC。(C)2009爱思唯尔B.V.保留所有权利。
The development of accurate and robust palmprint verification algorithms is a critical issue in automatic palmprint authentication systems. Among various palmprint verification approaches, the orientation based coding methods, such as competitive code (CompCode), palmprint orientation code (POC) and robust line orientation code (RLOC), are state-of-the-art ones. They extract and code the locally dominant orientation as features and could match the input palmprint in real-time and with high accuracy. However, using only one dominant orientation to represent a local region may lose some valuable information because there are cross lines in the palmprint. In this paper, we propose a novel feature extraction algorithm, namely binary orientation co-occurrence vector (BOCV), to represent multiple orientations for a local region. The BOCV can better describe the local orientation features and it is more robust to image rotation. Our experimental results on the public palmprint database show that the proposed BOCV outperforms the CompCode, POC and RLOC by reducing the equal error rate (EER) significantly. (C) 2009 Elsevier B.V. All rights reserved.