一种基于纹理特征点匹配的自适应虹膜识别方法  

Adaptive iris recognition method based on texture feature point matching

作　　者：苑玮琦[1] 林忠华[1] 徐露[1] 

机构地区：[1]沈阳工业大学视觉检测技术研究所,沈阳110023

出　　处：《仪器仪表学报》2009年第2期232-236,共5页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(60672078;60472088);辽宁省自然科学基金(20042028)资助项目

摘　　要：传统的虹膜识别方法原理复杂,实现困难。为了在保证识别率的同时简化虹膜识别算法,提高虹膜识别效率,提出了一种基于纹理特征点匹配的自适应虹膜识别方法。首先,通过图像预处理得到512列×64行的矩形虹膜图像;其次,在有效的虹膜区域内,通过自行设计的边缘检测方法找到虹膜特征点;再次,对特征点进行编码,用虹膜代码表示虹膜模式;最后,采用自适应的模式匹配方法对不同的虹膜模式进行分类,给出识别结果。在给定距离阈值为0.37的前提下,可以达到99.2428%的正确识别率。实验结果证明,该方法能够满足虹膜识别高准确性的要求。The principles of traditional iris recognition methods are complex and difficult to realize. In order to ensure the recognition rate, predigest the algorithm and improve the iris recognition efficiency, this paper presents an adaptive iris recognition method based on texture feature point matching. Firstly, a 512 columns by 64 rows rectangular iris image is obtained through image preprocessing. Secondly, iris feature points are obtained by segmentation using a detection template designed by ourselves in the effective iris area. Thirdly, the feature points are encoded and the iris codes are used to denote the iris pattern. Finally, different iris patterns are classified using adaptive pattern matching method and the recognition results are obtained. When the distance threshold equals to 0.37, the accuracy is 99. 242 8%. Many experiments demonstrate that this method can meet the high accuracy demand of iris recognition.

关 键 词：生物特征 虹膜识别 自适应 分块 检测模板 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]