摘要
提出了一种基于独立分量分析和支持向量机的虹膜识别方法—ICA提取虹膜特征,SVM实现模式匹配.与G abor小波的方法比较,在编码长度和编码时间方面有较明显地改进.实验结果表明,该算法能够有效地应用到身份鉴别系统中.
In this paper, a new iris recognition algorithm based on Independent Component Analysis and Support Vector Machine was proposed. ICA is applied to extract iris feature and SVM is used in pattern matching. Compared with Gabor wavelet method, the size of an iris code and the processing time of the feature extraction are significantly reduced. Experimental results show that the proposed system could be used for a personal identification system in an efficient and effective manner.
出处
《小型微型计算机系统》
CSCD
北大核心
2005年第12期2203-2206,共4页
Journal of Chinese Computer Systems
基金
湖南省自然科学基金项目(05JJ30123)资助
广东省自然科学基金重点项目(036608)资助
广州市科技计划项目(2003J1-C0201)资助.
关键词
虹膜识别
生物特征识别
独立分量分析
支持向量机
iris recognition
biometric
independent component analysis
support vector machine