請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/4029
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 林永松(Yeong-Sung Lin) | |
dc.contributor.author | Li-Min Zheng | en |
dc.contributor.author | 鄭立民 | zh_TW |
dc.date.accessioned | 2021-05-13T08:40:47Z | - |
dc.date.available | 2019-02-01 | |
dc.date.available | 2021-05-13T08:40:47Z | - |
dc.date.copyright | 2016-02-24 | |
dc.date.issued | 2016 | |
dc.date.submitted | 2016-01-27 | |
dc.identifier.citation | [1] Robert A. Mocny, Direcetor US-VISIT, “Biometrics Standards Requirements for US-Visit”
[2] Tracey Caldwell, “National ID cards in the UK: the role of biometrics”, Biometric Technology Today , Volume 2013, Issue 7, July 2013, Pages 7–8 [3] Bigthink editors, “World's Biggest Biometrics ID Scheme” [4] RNCOS E-Services Pvt. Ltd., Biometric Market Forecast to 2014 [5] Ariana-Michele Moore, “Biometric technologies — an introduction”, Biometric Technology Today, Volume 15, Issue 1, January 2007, Pages 6–7 [6] J.A.Unar , Woo Chaw Seng, Almas Abbasi, “A review of biometric technology along with trends and prospects”, Pattern Recognition, Volume 47, Issue 8, August 2014, Pages 2673-2688 [7] K. Bowyer, K. Chang, P. Flynn, X. Chen, “Face recognition using 2-D, 3-D, and infrared : is multi modal better than multi sample?”, Proc. IEEE’94, November 2006, Pages 2000–2012 [8] Biometrics.gov, the central source of information on biometrics-related activities of the Federal government, “Biometrics Technology and Standards Overview” [9] Norman Poh, Thirimachos Bourlai, Josef Kittler, “A multimodal biometric test bed for quality-dependent, cost-sensitive and client-specific score-level fusion algorithms”, Pattern Recognition, Volume 43, Issus 3, March 2010 ,Pages 1094-1105 [10] Amioy Kumar, M.Hanmandlu, H.M.Gupta, “Fuzzy binary decision tree for biometric based personal authentication”, Neurocomputing, Volume 99, 1 January 2013, Pages 87-97 [11] Y.J. Chin, T.S. Ong, A.B.J. Teoh, K.O.M. Goh, “Integrated biometrics template protection technique based on fingerprint and palmprint feature-level fusion” , Information Fusion Volume 18, July 2014, Pages 161–174 [12] NormanPoh , ArunRoss , WeifengLee , JosefKittler, “A user-specific and selective multimodal biometric fusion strategy by ranking subjects” , Pattern Recognition Volume 46, Issue 12, December 2013, Pages 3341–3357 [13] Anne M.P. Canuto, Fernando Pintro, Joao C. Xavier-Junior, “Investigating fusion approaches in multi-biometric cancellable recognition” , Expert Systems with Applications Volume 40, Issue 6, May 2013, Pages 1971–1980 [14] Suresh Kumar Ramachandran Nair, Bir Bhanu, Subir Ghosh, Ninad S. Thakoor, “Predictive models for multibiometric systems”, Pattern Recognition, Volume 47, Issue 12, December 2014, Pages 3779-3792 [15] Salman H.Khan, M. Ali Akbar, Farrukh Shahzad, Mudassar Farooq, Zeashan Khan, “Secure biometric template generation for multi-factor authentication”, Pattern Recognition, Volume 48, Issue 2, February 2015, Pages 458-473 [16] Convex Optimization (S.Boyd, L.Vandenberghe) Cambridge University Press, 2004 , ISBN-13:978-0521833783 , ISBN-10: 0521833787 [17] S. Boyd, S.-J. Kim, L. Vandenberghe, and A. Hassibi, “A Tutorial on Geometric Programming”, Optimization and Engineering, Volume 8, Issue 1, March 2007, Pages 67-127 [18] GGPLAB’, a matlab toolbox for specifying and solving geometric programs. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/4029 | - |
dc.description.abstract | 資訊安全近年成為企業組織十分重視之領域,透過網路處理任何資料型態,不論是結構化或是非結構化,log紀錄檔、照片、聲音、通訊紀錄或是電子郵件,包括使用者之隱私資訊,對該資訊之保護更是不可忽略。隱私資訊被竊取或濫用,對於商譽等無形資產之損害更是難以想像。例如:iCloud的名人照片被駭客攻擊竊取之事件,暴露出一般身份驗證機制不足,但是蘋果手機iPhone 5s以上其實已具備了多重辨識技術服務,包括人臉辨識、語音辨識、指紋辨識、或鍵盤輸入間隙辨識等生物認證技術,只是沒有一個良好生物辨識技術得以將其整合應用,故本研究為了貼近實際應用,擬提出多輪迴方式驗證方式,透過使用多重辨識技術融合(如帳號、密碼 、指紋 、瞳孔及人臉)之特徵值驗證防範不當連線。設計最佳化演算法達成一個高準確率及低誤判率的多重生物辨識認證系統,且對於驗證時間希望能在一定時間內完成,動態地依據安全性需求提供最佳的生物辨識組合方案,本計畫針對下列議題進行深入研究:
議題I:一個高準確率的多重生物辨識認證系統; 議題II:能夠動態的依據安全性需求提供最佳的生物辨識組合方案; 上述議題將運用數學式建立相關模型成目標式和多項限制式,依使用者只要在循序式的生物辨識機制的其中一個回合,能夠達到第一門檻即可通過,若是低於第二門檻則直接拒絕,若是位於其中,才會需要進行下一階段的生物辨識機制。如此一來,使用者只有在最糟糕的情形下才會需要使用所有系統所提供的生物辨識機制。計畫中除了運用數學模式來描述外,運用現有最佳化技術與自行開發優化演算法來進行分析和驗證,發展以幾何規劃為基礎的集中式與啟發式演算法,且執行相關參數驗證,依此設計一系列實驗、數值分析計算出最佳解或近似最佳解以最有效率及有效果之方式設計該多重生物辨識系統,結合理論與實務應用。 | zh_TW |
dc.description.abstract | In recent years, business organizations highly regarded in information security issues. We process any type of data whether structure or non-structure through the internet, such as log files, photographs, Voices, Communication records and e-mail. All these data has personal privacy information, so the protection of these data can not be ignore. The stealing and misuse of privacy information will harm company’s goodwill and the loss is hard to evaluate. For example, the hacker attack event in stealing celebrity photos on iCloud shows the insufficient of general identity authentication mechanism. Actually, iPhone 5s could provide several biometrics services include face recognition, voice recognition, fingerprint recognition and keystroke recognition, but there hasn’t exist a good way to intergrate all these biometrics services. Therefore, this paper propose a multi-modal biometrics authentication modal with serial verification in practical applications to avoid improper connection to the systems. This paper design an optimal algorithm to produce an optimal solution of biometrics combination with dynamic security requirements for a shortest authentication time, high-accuracy and low false reject multi-modal biometrics authentication systems. In our paper, we focus on the two issues:
Issues 1: A high-accuracy multi-modal biometrics system Issues 2: An optimal combination solution of biometric modality according to dynamic security requirements. Issues of above will use mathematical formula to establish the relevant model to the objective function and constraints. In our model, if user achieve the pass threshold, the system will accept the user; if user can not achieve the, the system will reject the user; only if user between pass threshold and reject threshold have to carry on the next level biometric authentication. User need to use all the biometric mechanism which provided by systems only in the worst-case situation. In addition to use mathematical formula to establish our model, we use exist optimization techniques and self-developed centralized and heuristic optimization algorithm based on Geometric Programming to analysis and verification. According to the design of series experiments and numerical analysis, we want to calculate optimal or near-optimal solutions in most efficiency and effectiveness way for the multi-modal biometrics systems which integrate theory with practice. | en |
dc.description.provenance | Made available in DSpace on 2021-05-13T08:40:47Z (GMT). No. of bitstreams: 1 ntu-105-R02725041-1.pdf: 1854132 bytes, checksum: d4e70d1df7567a089b5625d2ab98a0fd (MD5) Previous issue date: 2016 | en |
dc.description.tableofcontents | Contents
1. INTRODUCTION 1 1.1 BACKGROUND 1 1.2 MOTIVATION 2 1.3 RELATED WORK 3 1.3.1 Architecture of Biometric Systems 4 1.3.2 Modes of Operations 4 1.3.3 Peformance Measurement 6 1.3.4 Literature Survey 7 1.3.5 Operation Modes and Fusion Method of Multi-Modal Biometric Systems 8 2. PROBLEM FORMULATION 11 2.1 THE MULTI-MODAL BIOMETRICS MODOL WITH DOUBLE THRESHOLDS 11 2.2 PROBLEM DESCRIPTION 13 2.3 MATHEMATICAL FORMULATION 14 3. SOLUTION APPROACH 17 3.1 PERMUTATIONS AND COMBINATIONS OF BIOMETRICS 17 3.2 FINDING THE OPTIMAL SOLUTION WHICH CAN FULFILL THE FAR/FRR CONSTRAINTS 18 3.3 GEOMETRIC PROGRAMMING 18 3.3.1 GP modeling 19 3.3.2 Remodeling for the GP form 20 3.3.3 Solving the GP problem 21 4. COMPUTATIONAL EXPERIMENTS 24 4.1 EXPERIMENTAL ENVIROMENT 24 4.2 EXPERIMENTAL DATA 24 4.3 RESULT 25 4.3.1 Focus on authentication time(Experiment 4-1) 25 4.3.2 Focus on performance(Experiment 4-2,4-3) 27 4.3.3 Focus on permutation(Experiment 4-3,4-4) 28 5. FUTURE WORK 32 6. CONCLUSION 33 REFERENCE 34 | |
dc.language.iso | en | |
dc.title | 基於雙重門檻決策的循序型多重生物辨識認證系統 | zh_TW |
dc.title | A Serial Multi-modal Biometrics Authentication System Based on Double Threshold Decisions | en |
dc.type | Thesis | |
dc.date.schoolyear | 104-1 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 孔令傑(Ling-Jie Kong),鐘順平(Shun-Ping Jhong),呂俊賢(Jyun-Sian Lyu),莊東穎(Dong-Ying Jhuang) | |
dc.subject.keyword | 多重生物辨識系統,幾何規劃,雙重門檻,最佳化,循序生物辨識系統, | zh_TW |
dc.subject.keyword | Multi-Biometrics System,Geometric Programming,Double Thresholds,Optimization,Serial Biometrics System, | en |
dc.relation.page | 34 | |
dc.rights.note | 同意授權(全球公開) | |
dc.date.accepted | 2016-01-27 | |
dc.contributor.author-college | 管理學院 | zh_TW |
dc.contributor.author-dept | 資訊管理學研究所 | zh_TW |
顯示於系所單位: | 資訊管理學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-105-1.pdf | 1.81 MB | Adobe PDF | 檢視/開啟 |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。