作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (05): 176-179. doi: 10.3969/j.issn.1000-3428.2007.05.063

• 人工智能及识别技术 • 上一篇    下一篇

基于曲面匹配的虹膜识别方法

苑玮琦,高斌秀   

  1. (沈阳工业大学视觉检测技术研究所,沈阳 110023)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-03-05 发布日期:2007-03-05

Iris Recognition Method Based on Surface Match

YUAN Weiqi, GAO Binxiu   

  1. (Computer Vision Group, Shenyang University of Technology, Shenyang 110023)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-03-05 Published:2007-03-05

摘要: 提出了一种基于曲面匹配的虹膜识别方法,直接利用2幅虹膜图像对应的像素点灰度进行匹配。该方法主要包括虹膜图像尺寸和灰度范围的归一化、匹配区域的选择、匹配相似度阈值的估计以及对人眼旋转的自动校正。该文选择在眼毛、眼睑通常遮挡不到的一个较小的区域作为虹膜匹配区域,使该方法具有较宽的适应能力。没有特征提取过程,使得该方法具有计算量小的特点。图像之间的匹配过程使得伪造和窃取虹膜编码变得困难。实验表明该方法具有较高的识别率。

关键词: 虹膜识别, 安全性, 曲面匹配, 方差

Abstract: This paper proposes a method of the iris recognition based on the curved surface match, which carries out match by comparing the difference of the corresponding pixel gray between the two images. It mainly includes regularization of iris image dimension and gray, the selection of match region, the estimation of match similarity and auto condition of eyeball turn. The method has wider adaptability because of selecting a small region not to be kept out by eyelash and lid for iris match region. It has fast computing speed because there is no process of feature extraction. The match between images enables the forging filch of iris encoding to be difficult. The experiment result shows that the proposed method has high recognition rates.

Key words: Iris recognition, Security, Surface match, Variance