基于细节点描述子的指纹检索算法  

Fingerprint Indexing Based on Minutiae Descriptor

作　　者：汤正刚 沈雷[1] 吕葛梁 TANG Zheng-gang;SHEN Lei;LV Ge-liang(School of Communication Engineering,Hangzhou Dianzi University,Hangzhou 310018,China)

机构地区：[1]杭州电子科技大学通信工程学院,浙江杭州310018

出　　处：《软件导刊》2018年第12期100-103,107,共5页Software Guide

基　　金：中国博士后科学基金项目(2014M562302)

摘　　要：针对指纹一对一匹配识别方法严重影响指纹数据库系统识别效率的问题,提出一种基于细节点描述子的指纹检索算法,该算法先根据细节点描述子的结构信息进行粗匹配筛选,剔除虚假细节点产生的影响;然后利用筛选后的细节点位置关系确定最佳参考点;最后基于最佳参考点计算所有指纹图像相似度,将相似度排在前N0的指纹返回并作为候选指纹。实验结果表明,该算法在时间和空间上都优于传统指纹检索算法。Aiming at the problem that fingerprint one-to-one matching identification method seriously affects the recognition efficiency of fingerprint database system,a fingerprint retrieval algorithm based on minutiae descriptor is proposed.The algorithm first performs rough matching screening based on the structural information of the minutiae descriptors to remove the influence of false minutiae points,and then uses the filtered results of the location of the minutiae point to determine the best reference point.At last the algorithm calculates the similarity of the fingerprint images based on the best reference point,and selects the fingerprints with the similarity ranked in the top N0as the candidate fingerprints according to the the best reference point.Experimental results show that the proposed algorithm outperforms the traditional fingerprint indexing algorithm both in time and space.

关 键 词：指纹分类 指纹检索 细节点描述子 最佳参考点 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]