首页 | 本学科首页   官方微博 | 高级检索  
     

改进的基于人眼结构特征的虹膜识别方法
引用本文:苑玮琦,林忠华,徐露. 改进的基于人眼结构特征的虹膜识别方法[J]. 光电工程, 2007, 34(8): 105-109,133
作者姓名:苑玮琦  林忠华  徐露
作者单位:沈阳工业大学,视觉检测技术研究所,辽宁,沈阳,110023;沈阳工业大学,视觉检测技术研究所,辽宁,沈阳,110023;沈阳工业大学,视觉检测技术研究所,辽宁,沈阳,110023
基金项目:国家自然科学基金 , 辽宁省自然科学基金
摘    要:本文提出了一种改进的基于人眼结构特征的虹膜识别方法,基本思想是:首先,通过图像预处理,确定虹膜的位置和大小,将环形虹膜图像展开成矩形并进行归一化与图像增强;其次,利用局部灰度极小值的方法寻找有效虹膜区域内的特征点,依据纹理长度和方向信息去掉伪特征点,得到虹膜二进制编码;最后根据匹配准则进行识别.大量实验表明,该方法识别率高,识别时间短.

关 键 词:虹膜识别  图像预处理  特征提取  图像编码
文章编号:1003-501X(2007)08-0105-05
收稿时间:2006-10-09
修稿时间:2006-10-092007-05-16

Improved iris recognition method based on iris structure
YUAN Wei-qi,LIN Zhong-hua,XU Lu. Improved iris recognition method based on iris structure[J]. Opto-Electronic Engineering, 2007, 34(8): 105-109,133
Authors:YUAN Wei-qi  LIN Zhong-hua  XU Lu
Affiliation:Computer Vision Group, Shenyang University of Technology, Shenyang 110023, China
Abstract:Because of the unique structure characteristics and abundant information of irises, iris recognition has the highest accuracy in all of the personal identification technologies. An improved iris recognition algorithm is presented based on the structures of human eyes. Firstly, get the location and size of the iris through the image preprocessing, and then unwrap the localized iris to a rectangular texture block and make normalization and enhancement. Secondly, detect all of the feature points in the effective iris area utilizing the method of local gray minimum, and then abandon the false feature points by the information of texture length and direction and get a binary iris code, Finally, obtain the recognition result according to the matching rules.
Keywords:iris recognition   image preprocessing   feature extraction   image encoding
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号