毕业设计：指纹识别中若干关键问题研究.doc

摘要摘要还可以再修改一下。
随着信息时代的发展, 自动指纹识别技术已经广泛地应用于公安、海关、银行、网络安全等需要进行身份识别领域。本文概述了自动指纹识别系统的研究现状和指纹识别系统的算法流程,并在此基础上重点研究了指纹图像的分割算法和指纹图像细化算法。
首先,本文系统介绍了指纹识别的研究意义及现状,并对指纹识别原理及实现过程做了系统介绍。
其次,针对在指纹采集过程中存在着大量的低质量指纹图像,影响指纹识别系统的识别率的问题,本文对指纹图像的分割等预处理作了较为深入的研究,提出采用了均值方差的指纹图像分割算法,仿真结果表明,该方法有效的地改善了指纹图像的质量。
此外,对指纹图像的细化作了深入的研究,使用的细化模板虽然数量较多,计算时间比较长,但对于像素的处理很到位,使得毛刺比较少,保持原图像的基本结构特性,还最大地保留了细节特征。仿真结果表明。。。当然,鉴于此法速度慢,仍存在毛刺的缺点,还可以有很大的改进。
最后,研究了指纹识别过程中特征的提取方法,针对细节点提取过程中存在伪特征点的问题,本文提出采用了一种消除伪特征点的算法。仿真结果表明,这种算法可以有效的去除伪特征点。
关键词:指纹识别;均值方差;指纹细化;特征提取
ABSTRACT
With the developments of the information age, automated fingerprint identification technology has been widely used in public security, customs, banking. This article outlines both the study status and algorithm process of a Automated Fingerprint Recognition System, and on this basis, focuses on the segmentation of fingerprint image and fingerprint image thinning algorithm to do the study.
First, the paper describes the research significance of fingerprint recognition and its status. Fingerprint recognition theory and the key steps in the algorithm are introduced in this part .They are also used for programming verification.
Secondly, in the fingerprint collection process, there are existences of a large number of low-quality fingerprint images, affecting the identification rate of fingerprint recognition system problems. The paper focuses on fingerprint image segmentation which belongs to pre-research makes a more thorough study, putting forward a fingerprint image segmentation algorithm by used the mean and variance of the image. Calculating the mean and variance of each piece, and then if this result is almost close to 0 that it is regarded as the background. The area variance is not zero will use threshold segmentation algorithm.
In addition, the fingerprint image thinning is studied deeply, although the number of templates to use is a lot, calculation time is longer, the deal is in place for the pixels, making relatively few glitche