基于细节特征点的自动指纹识别算法与系统研究

发布时间：2018-06-16 15:40

本文选题：指纹识别 + 图像增强　；参考：《深圳大学》2016年硕士论文

【摘要】：随着计算机技术的迅猛发展,人们对身份识别技术的准确性、实用性以及安全性提出了越来越高的要求,基于生物特征的身份识别技术得到了广泛的关注。指纹特征具有唯一性和稳定性,以及指纹识别技术的高准确率、方便实用的特点,指纹识别技术已经成为目前最流行的个人身份识别技术之一,在信息安全、电子商务、智能手机、犯罪识别等领域得到了广泛应用。近年来,国内外学者对自动指纹识别技术进行了大量研究,自动指纹识别技术的研究已经成为学术界的热点,但是,指纹识别仍然存在不少的难题。本文分析了近些年来国内外指纹识别技术研究成果,在现有理论的基础上,用Visual C++实现了一套完整的指纹识别算法,并对其中的图像增强、图像细化、特征匹配等问题进行了深入的研究,并且提出了一些改进方法。本文的研究工作有以下几个方面:(1)对于断裂指纹的图像增强问题,本文提出了一种基于Gabor小波的智能收敛增强算法,利用Gabor小波的频向特性,使其在指纹纹线的方向场上对图像进行增强,以弥补图像中纹线的断裂等不足,在指纹纹线的垂直方向上,Gabor小波函数正好符合指纹纹线相间的特点,可以在该位置频率场上对图像纹线进行振荡加强,从而达到指纹图像增强的目的。(2)对OPTA(One-Pass Thinning Algorithm)细化算法进行了改进,采用统一的4?4模板作为改进的OPTA细化结构模板,改善了原OPTA细化算法中消除模板和保留模板不一致的情况,通过构建8个消除模板、6个保留模板实现对图像的细化,解决了原OPTA细化算法分叉点细化不彻底、纹线有大量毛刺产生的情况。最后,实验结果表明,达到了良好的效果。(3)本文在指纹特征匹配算法上做了改进,采取了一种基于脊线校正和极坐标相结合的匹配方法,引入了可变大小的界限盒,使算法更加支持非线性形变,增强了自动指纹识别系统的鲁棒性,提高了指纹的识别率和匹配速度。
[Abstract]:With the rapid development of computer technology, people have put forward more and more demands on the accuracy, practicality and security of identification technology. The identification technology based on biometrics has received extensive attention. The characteristics of fingerprint are unique and stable, as well as the high accuracy of finger recognition technology and convenient and practical features. Fingerprint recognition technology has become one of the most popular technology in personal identification. It has been widely used in the fields of information security, electronic commerce, smart phone, crime recognition and so on. In recent years, scholars at home and abroad have done a lot of research on automatic fingerprint recognition technology, and the research of auto fingerprint recognition technology has become a hot topic in the academic circle. However, there are still a lot of problems in fingerprint recognition. This paper analyzes the research results of fingerprint recognition technology at home and abroad in recent years. On the basis of the existing theory, a complete fingerprint recognition algorithm is realized with Visual C++, and the problems of image enhancement, image thinning and feature matching are studied in depth. There are some improvements in this paper: (1) for the problem of image enhancement of fractured fingerprint, an intelligent convergence enhancement algorithm based on Gabor wavelet is proposed in this paper, which makes use of the frequency characteristic of Gabor wavelet to enhance the image in the direction field of fingerprint lines to make up the broken lines in the image. In the vertical direction of fingerprint lines, the Gabor wavelet function coincides with the characteristics of the fingerprint lines, and can strengthen the image lines on the frequency field of the location, thus achieving the purpose of fingerprint image enhancement. (2) the refinement of the OPTA (One-Pass Thinning Algorithm) refinement algorithm is improved, and a unified 4? 4 template is adopted. As an improved OPTA thinning structure template, the original OPTA thinning algorithm can eliminate the inconsistency between the template and the reserved template. By constructing 8 elimination templates and 6 reserved templates to realize the thinning of the image, the original OPTA thinning algorithm is not completely refined, and the lines have a large number of burrs. Finally, the experimental result table It has achieved good results. (3) this paper improved the fingerprint feature matching algorithm, adopted a matching method based on the ridge line correction and the polar coordinates, introduced the variable size limit box, made the algorithm more nonlinear deformation, enhanced the robustness of the automatic finger recognition system, and improved the recognition rate of fingerprint. Match the speed.
【学位授予单位】：深圳大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.41

【相似文献】