欢迎访问《图学学报》 分享到：
图学学报

图学学报

• 视觉与图像 • 上一篇    下一篇

基于权值相位差的虹膜匹配算法

  

  1. 1. 上海理工大学光电信息与计算机工程学院，上海 200093；
    2. 上海师范大学信息与机电工程学院，上海 200234；
    3. 洛阳师范学院信息技术学院，河南 洛阳 471934
  • 出版日期:2017-04-30 发布日期:2017-04-28

An Iris Matching Algorithm Based on Weighted Phase Difference

  1. 1. School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
    2. College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China;
    3. College of Computer and Information, Luoyang Normal University, Luoyang Henan 471934, China
  • Online:2017-04-30 Published:2017-04-28
  • Supported by:
    国家自然科学基金项目(61502220, U1304616)；东方学者基金项目(15HJPY-MS02)；上海市自然科学基金项目(15ZR1428600)

PDF (PC)

188

摘要： 虹膜识别是最具发展前途的生物识别技术之一，由于虹膜匹配和评价作为虹膜识别
的核心步骤直接影响虹膜识别的准确率，提出一种新的基于权值相位差的虹膜匹配算法。首先对
定位好的虹膜图像重新采样，其次对其进行归一化处理，得到矩形虹膜图像。然后利用频域变换
处理矩形虹膜图像。最后采用新的虹膜匹配算法利用全部相位信息和部分振幅信息来进行虹膜的
匹配。虹膜纹理信息主要集中在虹膜环靠近瞳孔的位置，这部分区域纹理信息密集，本文提出的
基于权重函数的方法能经过实验验证，该算法有效地突出虹膜纹理低频信息部分快速准确，提高
了匹配效率，为虹膜识别的匹配和评价核心步骤提供了一个新的方法。

关键词: 虹膜识别, 相位差, 匹配算法, 生物识别技术

Abstract: Iris recognition technology is one of the most promising biological recognition technologies.
As the core procedure of iris recognition, matching and evaluation directly affect the accuracy of iris
recognition. In this paper, a novel iris matching algorithm based on the weighted phase difference is
proposed. The key steps of iris recognition are: First, resample the iris image which is well located.
Second, normalize the resampling points to obtain a rectangular iris image. Then, process the iris image
in Fourier space. Finally, the newly suggested algorithm utilizes all the phase information of the iris
image and partial amplitude information to do irises matching procedure. Iris texture information is
mainly concentrated to the pupil, where the texture information is rather intensive. Here, we propose a
novel method based on the weighted function of phase difference, which can effectively highlight the
intensive iris texture information in the low frequency part and improve the matching efficiency. The
experimental results show that the algorithm we propose improves greatly in accuracy and computation
speed, which is a good substitute for the core procedure of iris matching and evaluation.

Key words: iris recognition, phase difference, match algorithm, biological recognition technology