摘要
手形识别和掌纹识别是两种不同的生物特征识别技术,根据二者的特点,将它们相结合研究手形和掌纹融合算法。分别介绍了基于几何特征的手形特征提取方法和基于模糊方向能量的掌纹特征提取方法,针对两种生物特征的特点,重点研究了二者在决策层融合的方法,构建基于并联融合和串联融合的双模态识别系统。通过对每种算法的实验测试,表明融合后的系统性能要优于单一的手形识别和掌纹识别方法的性能。
Hand-geometry identification and palmprint identification are two kinds of biometric identification technology.A fusion algorithm fusing the hand-geometry and palmprint identification on the basis of one＇s unique physiological characteristics is presented.Respectively the feature extration method of hand-geometry based on geometrical characteristics and an approach of line feature representation and matching for palmprint recognition are introduced.According to the two biological features,the decision level fusion algorithm and set up a multi-modal system based on parallel connection and series connection are studied.From the experiment test toward each algorithm,it can be proved that the multi-modal system is stronger than single hand geometry and palmprint identify.
出处
《黑龙江大学工程学报》
2010年第4期122-127,共6页
Journal of Engineering of Heilongjiang University
基金
国家大学生科技训练计划项目(091021716)
关键词
手形识别
掌纹识别
双模态识别系统
决策层融合
hand-geometry identification
palmprint identification
multibiometric system
decision fusion
作者简介
王科俊（1962-），男，吉林吉林人，教授，博士研究生导师，主要研究方向：生物特征识别技术与智能监控，E-mail：wangkejun@hrbeu．edu．cn。
