本文選題：智能手機(jī) + 信息安全��； 參考：《吉林大學(xué)》2015年碩士論文

【摘要】：隨著移動(dòng)互聯(lián)網(wǎng)技術(shù)和智能手機(jī)終端的不斷完善，智能手機(jī)移動(dòng)支付受到了前所未有的關(guān)注。而隨著通信和信息化水平的提高，人們也越來(lái)越普遍地使用手機(jī)來(lái)存儲(chǔ)和傳輸大量的個(gè)人信息與隱私，并且希望保證個(gè)人信息安全。 指紋識(shí)別技術(shù)是目前應(yīng)用于智能手機(jī)終端的最為成熟的身份認(rèn)證與支付交易認(rèn)證技術(shù)，研究基于智能手機(jī)終端的指紋識(shí)別算法具有十分廣泛的重要意義。指紋識(shí)別效果的好壞，除了取決于指紋采集技術(shù)的高低，更在很大程度上取決于指紋識(shí)別算法的優(yōu)劣。預(yù)處理過(guò)程是指紋識(shí)別中重要的一環(huán)。所以在本文中，作者將著重針對(duì)預(yù)處理過(guò)程中涉及到的三個(gè)主要步驟進(jìn)行研究：圖像增強(qiáng)、二值化和細(xì)化。 我們會(huì)采用FVC2004（Fingerprint Verification Competition2004,2004指紋識(shí)別競(jìng)賽）的DB2和DB4指紋數(shù)據(jù)庫(kù)來(lái)對(duì)指紋進(jìn)行性能測(cè)試。經(jīng)FVC2004指紋性能測(cè)試的仿真試驗(yàn)驗(yàn)證，我們可以得出如下結(jié)論：基于相同的特征提取和匹配算法，本文提出的指紋圖像預(yù)處理優(yōu)化算法更加高效，能比傳統(tǒng)算法更好地消除噪音干擾，避免各種偽特征信息。 針對(duì)上述問(wèn)題本文擬開(kāi)展如下研究： 1.基于傅里葉和方向?yàn)V波的指紋圖像增強(qiáng)算法 基于Gabor濾波的增強(qiáng)算法，對(duì)較高質(zhì)量指紋圖像處理效果好，但占用預(yù)處理過(guò)程的時(shí)間長(zhǎng)�；诟道锶~濾波的增強(qiáng)算法，速度快并對(duì)低質(zhì)量指紋圖像的增強(qiáng)效果好。本文采用基于傅里葉變換和方向?yàn)V波的指紋圖像增強(qiáng)算法，實(shí)驗(yàn)證明能夠取得更好的指紋圖像處理效果。 2.基于方向圖場(chǎng)的動(dòng)態(tài)閾值二值化算法 基于固定閾值的二值化算法，雖然算法簡(jiǎn)單，但處理質(zhì)量差的指紋圖像會(huì)產(chǎn)生偽特征信息；基于動(dòng)態(tài)閾值的二值化算法，雖然在某些方面能夠彌補(bǔ)固定閾值算法的不足，，但也會(huì)引入噪音。本文采用基于圖場(chǎng)的動(dòng)態(tài)閾值算法，經(jīng)實(shí)驗(yàn)證明可以取得良好的抗噪性。 3.基于改進(jìn)的OPTA細(xì)化算法 快速細(xì)化算法利用像素的4領(lǐng)域判斷指紋邊界點(diǎn)并逐步刪除；OPTA細(xì)化算法利用8個(gè)消除模板和2個(gè)保留模板進(jìn)行對(duì)比，判斷中心像素是否刪除；改進(jìn)的OPTA細(xì)化算法改造了算法的消除模板和保留模板。本文將在進(jìn)行改進(jìn)的OPTA細(xì)化算法的基礎(chǔ)上，對(duì)細(xì)化后的指紋圖像采用方形模板進(jìn)行紋線跟蹤，實(shí)驗(yàn)證明與傳統(tǒng)算法相比較，本文提出的算法能夠更好地去除偽特征點(diǎn)、克服干擾因素。
[Abstract]:With the continuous improvement of mobile Internet technology and smart phone terminals, mobile payment has received unprecedented attention. With the improvement of communication and information technology, mobile phones are more and more widely used to store and transmit a large amount of personal information and privacy, and they hope to ensure the security of personal information. Fingerprint identification technology is the most mature authentication and payment authentication technology used in smart phone terminal. It is very important to study fingerprint identification algorithm based on smart phone terminal. The effect of fingerprint recognition depends not only on the level of fingerprint acquisition technology, but also on the advantage and disadvantage of fingerprint recognition algorithm to a great extent. Pretreatment is an important part of fingerprint recognition. Therefore, in this paper, the author will focus on three main steps involved in the process of preprocessing: image enhancement, binarization and thinning. We will use the DB2 and DB4 fingerprint databases of FVC2004(Fingerprint Verification Competition 2004 to test fingerprint performance. Through the simulation of FVC2004 fingerprint performance test, we can draw the following conclusions: based on the same feature extraction and matching algorithm, the fingerprint image preprocessing optimization algorithm proposed in this paper is more efficient. It can eliminate noise interference and avoid all kinds of pseudo-feature information better than traditional algorithm. In view of the above problems, this paper intends to carry out the following research: 1. Fingerprint Image Enhancement algorithm based on Fourier and Direction filtering The enhancement algorithm based on Gabor filter has good effect on fingerprint image processing with high quality, but it takes up a long time in the process of preprocessing. The enhancement algorithm based on Fourier filter has high speed and good enhancement effect on low quality fingerprint image. In this paper, a fingerprint image enhancement algorithm based on Fourier transform and directional filtering is proposed, and the experimental results show that it can achieve better fingerprint image processing results. 2. Dynamic threshold binarization algorithm based on pattern Field Although the binarization algorithm based on fixed threshold is simple, it can produce pseudo-feature information when dealing with bad quality fingerprint image, while the binarization algorithm based on dynamic threshold can make up for the deficiency of fixed threshold algorithm in some aspects. But it also introduces noise. The dynamic threshold algorithm based on graph field is used in this paper. 3. Improved OPTA thinning algorithm The fast thinning algorithm uses the four fields of pixels to judge the fingerprint boundary points and gradually deletes the OPTA thinning algorithm compared with 8 elimination templates and 2 reserved templates to judge whether the center pixels are deleted or not. The improved OPTA thinning algorithm has modified the elimination template and reserved template of the algorithm. Based on the improved OPTA thinning algorithm, this paper uses the square template to track the lines of the thinned fingerprint image. The experimental results show that the algorithm proposed in this paper can remove the pseudo-feature points better than the traditional algorithm. Overcome interference factors.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TP391.41;TN929.53

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王芳;;生物防偽技術(shù)及其應(yīng)用[J];中國(guó)防偽報(bào)道;2011年07期

2 陳穎;;對(duì)移動(dòng)支付發(fā)展的探討[J];長(zhǎng)沙通信職業(yè)技術(shù)學(xué)院學(xué)報(bào);2013年01期

3 陸建華;;移動(dòng)支付在鐵路客票系統(tǒng)中的應(yīng)用研究[J];高速鐵路技術(shù);2013年02期

4 商慶瑞;吳晴;郭晶瑩;;基于Matlab實(shí)現(xiàn)的指紋圖像預(yù)處理算法[J];計(jì)算機(jī)仿真;2007年03期

5 劉春明;;指紋特征規(guī)模研究[J];計(jì)算機(jī)仿真;2007年08期

6 盧洋;馮祥勝;;基于方向信息的指紋圖像的預(yù)處理與識(shí)別[J];計(jì)算機(jī)與現(xiàn)代化;2006年12期

7 馬鴻雁;;指紋識(shí)別原理及其圖像壓縮評(píng)價(jià)分析[J];甘肅科技縱橫;2008年05期

8 陳鋼;;指紋識(shí)別關(guān)鍵算法的研究[J];信息通信;2013年06期

9 黃世龍;劉書剛;閻嘉t

本文編號(hào)：1991787