自动指纹识别系统的研究与实现  

Research and Realization of Automatic Fingerprint Identification System

作　　者：王娜[1] WANG Na(Department of Information Engineering, LiaoNing Construction Vocational College, Liaoyang, 111000, China)

机构地区：[1]辽宁建筑职业学院信息工程系,辽宁辽阳111000

出　　处：《电脑与信息技术》2016年第5期7-9,21,共4页Computer and Information Technology

摘　　要：对指纹图像的预处理、特征提取和指纹匹配的算法进行了研究。在指纹图像预处理中,首先选取了一种简单的分割算法,即在获取指纹频率场和方向图前就可以完成,从而能够尽早获得有效区域;在脊线频率和方向图的基础上,实现了一种基于脊线方向分析的二值化算法;实现了利用数学形态学对二值化后的图像进行细化处理,构成了一个比较完整的预处理过程。在指纹图像特征提取和匹配中,每个指纹特征点的信息都以向量进行表示,向量构成了指纹的所有信息;为了提高指纹旋转后的匹配率,改进了指纹的特征提取和匹配的算法。最后,对研究的所有算法进行了测试,结果表明,利用这些算法能够很好的实现各自的功能,完成了预期的目标。Algorithms of the preprocessing, feature extraction and fingerprint matching is studied.In fingerprint image preprocessing, first select an easy segmentation method, namely before obtain fingerprint spectrum field and the pattern can be completed, as soon as possible so that they can get effective area; Implements a binarization algorithm based on ridge line direction analysis;Realize the use of mathematical morphology for thinning after binary image, formed a relatively complete preprocessing.In fingerprint image feature extraction and matching, information of each fingerprint feature are expressed as a vector, vector constituted all fingerprint information; In order to improve the fingerprint matching rate after the rotation, improved the algorithm of fingerprint feature extraction and matching.Finally, the study of all the algorithms were tested, the results show that the use of these algorithms can achieve good respective function,completed the expected goal.

关 键 词：自动指纹识别系统 预处理 特征提取 指纹匹配 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]