基于中心区域基准点的指纹匹配算法  

A FINGERPRINT MATCHING ALGORITHM BASED ON REFERENCE POINT OF CENTRAL AREA

作　　者：刘健[1] 王万良[1] 

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310023

出　　处：《计算机应用与软件》2011年第2期30-33,共4页Computer Applications and Software

基　　金：国家自然科学基金项目(60573123)

摘　　要：指纹匹配是自动指纹识别中的重要环节。首先介绍了指纹匹配中已有的一些方法和指纹匹配需要解决的问题,继而提出了一种基于中心区域基准点的指纹匹配算法,通过搜寻中心区域的细节点集合定位基准点对,以此计算旋转和平移参数,最后统计姿势矫正后的匹配点个数来实现两枚指纹的比对。实验结果表明,针对图像质量不同的两个指纹库,该算法能有效地抵抗图像噪声,准确定位基准点,达到较高的准确率。Fingerprint matching is an important issue in automatic fingerprint identification system.In this paper we describe some fingerprint matching algorithms that have been delivered and some problems of them to be resolved,and then propose a fingerprint matching algorithm based on reference points of the central area.The pair of the reference points can be located by scanning two minutia point sets in central area,and the parameters of rotation and translation are calculated accordingly.Finally,the intercomparison between two fingerprints is implemented by counting the number of matching points after the posture of image has been rectified.Experimental results indicate that for two fingerprint databases with different image qualities,this algorithm can effectively resist image noise,locate the reference point precisely and achieve higher accuracy rate.

关 键 词：基准点 细节点 匹配 中心 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]