基于細(xì)節(jié)特征點(diǎn)的自動(dòng)指紋識(shí)別算法與系統(tǒng)研究
本文選題:指紋識(shí)別 + 圖像增強(qiáng) ; 參考:《深圳大學(xué)》2016年碩士論文
【摘要】:隨著計(jì)算機(jī)技術(shù)的迅猛發(fā)展,人們對(duì)身份識(shí)別技術(shù)的準(zhǔn)確性、實(shí)用性以及安全性提出了越來(lái)越高的要求,基于生物特征的身份識(shí)別技術(shù)得到了廣泛的關(guān)注。指紋特征具有唯一性和穩(wěn)定性,以及指紋識(shí)別技術(shù)的高準(zhǔn)確率、方便實(shí)用的特點(diǎn),指紋識(shí)別技術(shù)已經(jīng)成為目前最流行的個(gè)人身份識(shí)別技術(shù)之一,在信息安全、電子商務(wù)、智能手機(jī)、犯罪識(shí)別等領(lǐng)域得到了廣泛應(yīng)用。近年來(lái),國(guó)內(nèi)外學(xué)者對(duì)自動(dòng)指紋識(shí)別技術(shù)進(jìn)行了大量研究,自動(dòng)指紋識(shí)別技術(shù)的研究已經(jīng)成為學(xué)術(shù)界的熱點(diǎn),但是,指紋識(shí)別仍然存在不少的難題。本文分析了近些年來(lái)國(guó)內(nèi)外指紋識(shí)別技術(shù)研究成果,在現(xiàn)有理論的基礎(chǔ)上,用Visual C++實(shí)現(xiàn)了一套完整的指紋識(shí)別算法,并對(duì)其中的圖像增強(qiáng)、圖像細(xì)化、特征匹配等問(wèn)題進(jìn)行了深入的研究,并且提出了一些改進(jìn)方法。本文的研究工作有以下幾個(gè)方面:(1)對(duì)于斷裂指紋的圖像增強(qiáng)問(wèn)題,本文提出了一種基于Gabor小波的智能收斂增強(qiáng)算法,利用Gabor小波的頻向特性,使其在指紋紋線的方向場(chǎng)上對(duì)圖像進(jìn)行增強(qiáng),以彌補(bǔ)圖像中紋線的斷裂等不足,在指紋紋線的垂直方向上,Gabor小波函數(shù)正好符合指紋紋線相間的特點(diǎn),可以在該位置頻率場(chǎng)上對(duì)圖像紋線進(jìn)行振蕩加強(qiáng),從而達(dá)到指紋圖像增強(qiáng)的目的。(2)對(duì)OPTA(One-Pass Thinning Algorithm)細(xì)化算法進(jìn)行了改進(jìn),采用統(tǒng)一的4?4模板作為改進(jìn)的OPTA細(xì)化結(jié)構(gòu)模板,改善了原OPTA細(xì)化算法中消除模板和保留模板不一致的情況,通過(guò)構(gòu)建8個(gè)消除模板、6個(gè)保留模板實(shí)現(xiàn)對(duì)圖像的細(xì)化,解決了原OPTA細(xì)化算法分叉點(diǎn)細(xì)化不徹底、紋線有大量毛刺產(chǎn)生的情況。最后,實(shí)驗(yàn)結(jié)果表明,達(dá)到了良好的效果。(3)本文在指紋特征匹配算法上做了改進(jìn),采取了一種基于脊線校正和極坐標(biāo)相結(jié)合的匹配方法,引入了可變大小的界限盒,使算法更加支持非線性形變,增強(qiáng)了自動(dòng)指紋識(shí)別系統(tǒng)的魯棒性,提高了指紋的識(shí)別率和匹配速度。
[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.
【學(xué)位授予單位】:深圳大學(xué)
【學(xué)位級(jí)別】:碩士
【學(xué)位授予年份】:2016
【分類號(hào)】:TP391.41
【相似文獻(xiàn)】
相關(guān)期刊論文 前7條
1 ;自動(dòng)指紋識(shí)別底層核心技術(shù)[J];技術(shù)與市場(chǎng);2005年12期
2 王植,方艷梅;自動(dòng)指紋識(shí)別在網(wǎng)絡(luò)安全中的應(yīng)用[J];現(xiàn)代計(jì)算機(jī);2000年10期
3 趙磊;蔣澎濤;劉春雷;陳宏君;楊夢(mèng)婷;;一種實(shí)用的自動(dòng)指紋識(shí)別完整算法[J];微計(jì)算機(jī)信息;2012年06期
4 王超峰;;嵌入式自動(dòng)指紋識(shí)別系統(tǒng)設(shè)計(jì)[J];微計(jì)算機(jī)信息;2011年07期
5 江杰;李杰;胡曉莉;;自動(dòng)指紋識(shí)別中的圖像預(yù)處理算法研究[J];微計(jì)算機(jī)信息;2005年26期
6 郭全花,李鳳云;基于網(wǎng)絡(luò)的自動(dòng)指紋識(shí)別系統(tǒng)[J];河北建筑工程學(xué)院學(xué)報(bào);2004年03期
7 ;[J];;年期
相關(guān)會(huì)議論文 前1條
1 宋曉玲;劉磊;;基于數(shù)據(jù)挖掘的自動(dòng)指紋識(shí)別系統(tǒng)設(shè)計(jì)[A];第十屆中國(guó)科協(xié)年會(huì)信息化與社會(huì)發(fā)展學(xué)術(shù)討論會(huì)分會(huì)場(chǎng)論文集[C];2008年
相關(guān)重要報(bào)紙文章 前1條
1 盧家興;中科院頻傳科研捷報(bào)[N];大眾科技報(bào);2004年
相關(guān)博士學(xué)位論文 前1條
1 任春曉;自動(dòng)指紋識(shí)別中若干關(guān)鍵算法的研究[D];山東大學(xué);2011年
相關(guān)碩士學(xué)位論文 前10條
1 任向飛;基于細(xì)節(jié)特征點(diǎn)的自動(dòng)指紋識(shí)別算法與系統(tǒng)研究[D];深圳大學(xué);2016年
2 趙振;自動(dòng)指紋識(shí)別系統(tǒng)算法的研究與實(shí)現(xiàn)[D];山東科技大學(xué);2009年
3 潘雷;自動(dòng)指紋識(shí)別原型系統(tǒng)的研究與實(shí)現(xiàn)[D];北京交通大學(xué);2008年
4 顧菘;基于DSP自動(dòng)指紋識(shí)別系統(tǒng)硬件平臺(tái)的設(shè)計(jì)與實(shí)現(xiàn)[D];電子科技大學(xué);2005年
5 李俊偉;自動(dòng)指紋識(shí)別關(guān)鍵算法的研究[D];青島理工大學(xué);2011年
6 徐俊峰;自動(dòng)指紋識(shí)別系統(tǒng)算法研究[D];西安電子科技大學(xué);2013年
7 梁珊;自動(dòng)指紋識(shí)別系統(tǒng)技術(shù)研究[D];中南大學(xué);2011年
8 朱弘;自動(dòng)指紋識(shí)別系統(tǒng)關(guān)鍵技術(shù)的研究與設(shè)計(jì)[D];合肥工業(yè)大學(xué);2006年
9 鄒軍;自動(dòng)指紋識(shí)別算法分析與設(shè)計(jì)[D];揚(yáng)州大學(xué);2008年
10 袁琳琳;自動(dòng)指紋識(shí)別方法的研究[D];華北電力大學(xué)(河北);2009年
,本文編號(hào):2027201
本文鏈接:http://sikaile.net/jingjilunwen/dianzishangwulunwen/2027201.html