基于相对距离的手指静脉识别方法  

Finger Vein Recognition Algorithm Based on Relative Distance

作　　者：王科俊[1] 刘靖宇[1] 李雪峰[1] 

机构地区：[1]哈尔滨工程大学自动化学院,哈尔滨150001

出　　处：《计算机科学》2011年第9期257-259,270,共4页Computer Science

基　　金：国家863高技术研究发展计划(2008AA01Z148);国家自然科学基金(60975022);黑龙江省杰出青年科学基金(JC200703);哈尔滨市科技创新人才研究专项基金(2007RFXXG009)资助

摘　　要：提出了一种新的手指静脉识别方法,即以静脉内部拓扑结构的本质特性为原理进行手指静脉的匹配,首先对细化修复后的图像提取端点和交叉点,并计算这些特征点之间的相对距离,最后通过比对这些距离值来完成手指静脉图像的识别。该方法结合静脉自身特征,充分利用了拓扑结构的本质属性,无须定位,简单易行。实验结果表明,该方法能够快速准确地进行身份识别,在一定程度上克服了平移、旋转对识别结果的影响,具有实际应用价值。Raised a new finger vein recognition algorithm,the method uses the essential feature of inner finger vein feature structure to match the finger vein.First,extracts the endpoint and crosspoint in the finger vein images which have been thinned and repaired.Then computes the distance between these points.Lastly accomplishes the finger vein image recognition by comparing these distance value.The method integrates the vein＇s inherent feature and makes full use of the essential property of topology structure,and the localization is needless,so the method is simple and easily to be executed.The experiment result indicates that the algorithm effectivly overcomes the influence of image translation and rotation,and the identity recognition can be executed fast and exactly.So the method has a actual applicative value and an evolvable forground.

关 键 词：手指静脉识别 拓扑结构 端点和交叉点 相对距离 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]