本文選題：人臉識(shí)別　切入點(diǎn)：ATM認(rèn)證　出處：《華南農(nóng)業(yè)大學(xué)》2016年碩士論文

【摘要】：自動(dòng)柜員機(jī),又稱ATM,是一種利用磁卡或智能卡介質(zhì)的方式實(shí)現(xiàn)金融交易的自助服務(wù)設(shè)備。客戶通過ATM做交易時(shí),需先將持有的磁卡或智能卡插入到ATM機(jī)中,ATM讀取介質(zhì)數(shù)據(jù)之后提示客戶輸入介質(zhì)密碼,系統(tǒng)認(rèn)證通過客戶輸入的密碼后才可進(jìn)行交易。當(dāng)前ATM此種認(rèn)證方式存在安全性差、易偽造、易被竊取等缺點(diǎn)。人臉識(shí)別技術(shù)的出現(xiàn),是解決傳統(tǒng)認(rèn)證方式缺點(diǎn)的方法。它具有防偽性強(qiáng)、不易偽造、特征唯一性等優(yōu)點(diǎn),然而當(dāng)前極少有銀行把人臉識(shí)別技術(shù)應(yīng)用到ATM系統(tǒng)中。本課題的任務(wù)是以ATM機(jī)認(rèn)證系統(tǒng)為研究對(duì)象,將人臉識(shí)別技術(shù)應(yīng)用到ATM機(jī)的認(rèn)證系統(tǒng)上,并且設(shè)計(jì)一套網(wǎng)絡(luò)環(huán)境下高負(fù)載、高可用、通用的人臉識(shí)別認(rèn)證系統(tǒng)解決方案。以此來(lái)解決當(dāng)前ATM機(jī)認(rèn)證系統(tǒng)遇到的問題。其中主要需要解決的問題是如何將人臉識(shí)別技術(shù)應(yīng)用到ATM系統(tǒng)中、如何獲取客戶的身份證標(biāo)準(zhǔn)照然后用于系統(tǒng)進(jìn)行人臉識(shí)別匹配、人臉識(shí)別匹配完成后如何發(fā)送彩信告知客戶識(shí)別匹配的結(jié)果及操作現(xiàn)場(chǎng)、如何支持認(rèn)證系統(tǒng)高并發(fā)訪問、負(fù)載均衡與災(zāi)備機(jī)制的需求。本論文開篇先闡述課題的研究背景以及意義,然后對(duì)國(guó)內(nèi)外人臉識(shí)別技術(shù)當(dāng)前現(xiàn)狀及發(fā)展進(jìn)行分析,接著闡述把人臉識(shí)別技術(shù)應(yīng)用到ATM系統(tǒng)中可行性的理論依據(jù)。之后介紹人臉識(shí)別技術(shù)的概念、技術(shù)原理以及人臉識(shí)別技術(shù)的特色與優(yōu)勢(shì)。接著對(duì)本論文研究?jī)?nèi)容進(jìn)行功能性需求分析及非功能性需求分析,然后介紹人臉識(shí)別認(rèn)證系統(tǒng)的總體架構(gòu)。另外,介紹人臉識(shí)別認(rèn)證系統(tǒng)需要集成的人臉識(shí)別攝像頭、短信平臺(tái)外部設(shè)備。完成系統(tǒng)架構(gòu)及外部設(shè)備介紹之后開始逐步構(gòu)建系統(tǒng)。其中包括API樁的設(shè)計(jì)與實(shí)現(xiàn),構(gòu)建內(nèi)部通信模塊、外設(shè)服務(wù)模塊、流程控制模塊、接口封裝;服務(wù)端的設(shè)計(jì)與實(shí)現(xiàn),構(gòu)建通信模塊、彩信中心模塊、數(shù)據(jù)庫(kù)訪問模塊、第三方系統(tǒng)訪問模塊、主流程控制模塊;路由控制端的設(shè)計(jì)與實(shí)現(xiàn),構(gòu)建內(nèi)部通信模塊、路由規(guī)則;數(shù)據(jù)庫(kù)表設(shè)計(jì);ATM認(rèn)證系統(tǒng)改造;最后基于路由控制實(shí)現(xiàn)人臉識(shí)別認(rèn)證系統(tǒng)的負(fù)載均衡與災(zāi)備機(jī)制方案。完成人臉識(shí)別認(rèn)證系統(tǒng)構(gòu)建后,通過用于自動(dòng)化測(cè)試與性能測(cè)試的軟件工具,利用測(cè)試軟件對(duì)人臉識(shí)別認(rèn)證系統(tǒng)進(jìn)行壓力測(cè)試,根據(jù)測(cè)試結(jié)果分析系統(tǒng)負(fù)載均衡與災(zāi)備機(jī)制是否有效,人臉識(shí)別認(rèn)證系統(tǒng)的性能可否滿足高并發(fā)的業(yè)務(wù)量。
[Abstract]:ATM (automatic teller machine) is a kind of self-service equipment which realizes financial transaction by means of magnetic card or smart card medium.When a customer makes a transaction through ATM, he must insert the magnetic card or smart card into the ATM to read the medium data and prompt the customer to enter the medium password, and the system authenticates through the password entered by the customer before the transaction can be carried out.Current ATM this kind of authentication way has the shortcoming such as the security, is easy to forge, is easy to be stolen and so on.The appearance of face recognition technology is a method to solve the shortcomings of traditional authentication methods.It has the advantages of strong anti-counterfeiting, hard to forge, unique features and so on. However, at present, very few banks apply face recognition technology to ATM system.The task of this paper is to apply face recognition technology to ATM authentication system, and to design a set of high load, high availability and general face recognition authentication system solution under network environment.In order to solve the current ATM authentication system encountered problems.The main problem that needs to be solved is how to apply face recognition technology to ATM system, how to obtain the standard photos of the customer's ID card and then apply it to face recognition matching.How to send MMS to inform customers of the matching result and operation site after face recognition matching is completed, and how to support the requirements of authentication system with high concurrent access, load balancing and disaster preparedness mechanism.At the beginning of this paper, the research background and significance of the subject are described, then the present situation and development of face recognition technology at home and abroad are analyzed, and the theoretical basis for the feasibility of applying face recognition technology to ATM system is expounded.Then it introduces the concept and principle of face recognition technology, as well as the features and advantages of face recognition technology.Then the functional requirements analysis and non-functional requirements analysis are carried out, and then the overall framework of face recognition authentication system is introduced.In addition, the face recognition authentication system needs the integrated face recognition camera and the external equipment of the short message platform.After completing the introduction of the system architecture and external equipment, we began to build the system step by step.It includes the design and realization of API pile, the construction of internal communication module, peripheral service module, process control module and interface encapsulation, the design and implementation of service end, the construction of communication module, MMS center module, database access module,Third party system access module, main process control module, design and implementation of routing control terminal, construction of internal communication module, routing rules, database table design and ATM authentication system transformation;Finally, the load balancing and disaster preparedness scheme of face recognition authentication system based on routing control is implemented.After completing the construction of the face recognition authentication system, through the software tools used for automated testing and performance testing, the test software is used to carry out the stress test on the face recognition authentication system.According to the test results, this paper analyzes whether the load balancing and disaster preparedness mechanism of the system is effective, and whether the performance of the face recognition authentication system can satisfy the high concurrent traffic.
【學(xué)位授予單位】：華南農(nóng)業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：TH693.5;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 Hai-Shun Du;Qing-Pu Hu;Dian-Feng Qiao;Ioannis Pitas;;Robust Face Recognition via Low-rank Sparse Representation-based Classification[J];International Journal of Automation and Computing;2015年06期

2 陳奇毅;;淺析人臉識(shí)別技術(shù)的現(xiàn)狀和發(fā)展趨勢(shì)[J];通訊世界;2015年13期

3 方程;;人臉識(shí)別技術(shù)研究[J];信息技術(shù)與信息化;2014年11期

4 Huan-Li Xu;Xin Zhao;Ke-Ming Zhang;Wei Tang;Norihiro Kokudo;;Inhibition of KL-6/MUC1 glycosylation limits aggressive progression of pancreatic cancer[J];World Journal of Gastroenterology;2014年34期

5 汪思成;;人臉識(shí)別技術(shù)的原理及應(yīng)用[J];電子技術(shù)與軟件工程;2014年09期

6 LI YueLong;MENG Li;FENG JuFu;WU JiGang;;Downsampling sparse representation and discriminant information aided occluded face recognition[J];Science China(Information Sciences);2014年03期

7 毛巨勇;;生物識(shí)別市場(chǎng) 大變革即將來(lái)臨——2013年生物識(shí)別市場(chǎng)發(fā)展回顧與熱點(diǎn)分析[J];中國(guó)安防;2013年12期

8 Sun Jiping;Li Chenxin;;In-pit coal mine personnel uniqueness detection technology based on personnel positioning and face recognition[J];International Journal of Mining Science and Technology;2013年03期

9 ;Novel Method of Mining Classification Information for SVM Training[J];Wuhan University Journal of Natural Sciences;2011年06期

10 董琳;趙懷勛;;人臉識(shí)別技術(shù)的研究現(xiàn)狀與展望[J];安防科技;2011年10期

，

本文編號(hào)：1719322