簡易檢索 / 詳目顯示

研究生: 黃志雄
David - Mulyono
論文名稱: 用於人員識別之手指靜脈的研究
A Study of Finger Vein Biometric for Personal Identification
指導教授: 洪西進
Shi-Jinn Horng
口試委員: 鍾國亮
Chung Kuo-Liang
鄭有進
Cheng You-Jinn
梅興
Mei-Xing
王振興
Jenn-Shing Wang
學位類別: 碩士
Master
系所名稱: 電資學院 - 資訊工程系
Department of Computer Science and Information Engineering
論文出版年: 2008
畢業學年度: 96
語文別: 英文
論文頁數: 125
中文關鍵詞: 人員識別手指靜脈生物特徵圖樣辨識適當的門檻值樣板匹配
外文關鍵詞: personal identification, finger vein biometric, pattern recognition, adaptive threshold, template matching
相關次數: 點閱:236下載:4
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

手指靜脈對於每個人是獨一無二的特徵,所以手指靜脈的驗證就安全性和便利性而言,在當下是主要的一項生物辨識技術。在紅外線光的照射下,利用攝影機抓取手指的圖像,其中除了包含指靜脈圖像外,在陰暗處也存在著手指的肌肉、骨頭和組織。在這篇論文中,介紹了初步的程序來提升因攝影機產生雜訊而被降低的圖像品質,並且使用適當的門檻值除去背景不相關的資訊,來得到清晰的指靜脈圖像。最後,使用改進過的樣板匹配方法來做指靜脈辨識。實驗結果顯示出即使圖像品質並不夠好,但只要指靜脈圖像是清晰的,利用一些適當的方法來處理,便可以運用在人員識別方面。因此,它依然可達到100%的識別正確率。


Finger vein authentication can be a leading biometric technology nowadays in terms of security and convenience, since it introduces the features inside the human body. An image of a finger captured by the camera under the IR light transmission contains not only the vein pattern itself, but also shade produced by various thickness of the finger muscles, bones, and tissue networks surrounding the vein. In this paper, we introduce preliminary process to enhance the image quality worsened by light effect and noises produced by the camera and then segment the vein pattern by using adaptive threshold method. Finally, matched them using improved template matching. The experimental result shows that even the image quality is not good enough, but as long as our veins are clear, processed by using the methods in this paper, it still can be used as the means of personal identification. Hence it still can achieve up to 100% identification accuracy when we take our 1000 images in our database in to the matching scheme.

摘要……………………………………………………………………..i Abstract…………………………………………….……………………..ii Acknowledgements…………………………………….………………….iii Table of Contents………………………………….……...........iv List of Figures ……………………………………………………..vii List of Tables………………………………………….............ix List of Equations………………………………….………………………x CHAPTER I. INTRODUCTION....................................1 I.1.Overview of Biometric Technologies……………………1 I.2.Motivation…………………………………………..6 I.3.Thesis Organization…………………………………7 CHAPTER II. SYSTEM ARCHITECTURE…………………………….9 II.1.Software Specification……………………………….9 II.2.Hardware Specification………………………………..10 •Storage Media ………………………………………..10 •Memory…………………………………………………………10 •Processor……………………………………………………….10 •Infrared…………………………………………………………10 •Camera………………………………………………………….14 •Filter……………………………………………………….14 •Finger vein grabber device………………………………..15 •Object detection………………………………………………16 •Object characteristics………………………………...18 •Process flow diagram………………………………...19 CHAPTER III. PREPROCESSING………………………………………20 III.1.Noise Removal…………………………………………...20 III.2.Dispelling the Illumination………………………………21 III.3.Image Normalization……………………………….22 CHAPTER IV. FEATURE EXTRACTION……………………………...24 IV.1.Adaptive Threshold………………………………………….24 IV.2.Median Filtering……………………………………………..26 IV.3.Massive Noises Removal…………………………………26 CHAPTER V. MATCHING……………………………………………30 V.1.Spatial Reduction…………………………………………….30 V.2.Re-labeling the Vein Image………………………………….30 V.3.Matching of Data……………………………………………31 CHAPTER VI. EXPERIMENTAL RESULT…………………………….34 VI.1.Database Generation…………………………………34 VI.2.Experimental Result…………………………………35 A.Complete-matching………………………………………………36 B.Self-matching……………………………………………36 C.Different-matching……………………………………………37 D.Threshold Accuracy……………..……………………………37 E.Intruder Access against Rm Threshold.…………………….40 F.Combination of Adaptive Threshold and Median Filter…41 G.Response time………………………………………………42 CHAPTER VII. CONCLUSION…………………………………………44 REFERENCES………………………………………………………46 CREDIT…………………………………………………………………..48 APPENDIX A (OPERATION EXAMPLES)……………………………49 APPENDIX B (COMPLETE MACTHING RESULT….……………….....53 APPENDIX C (FINGER IMAGE DATABASE)………………………..74

[1]Naoto Miura, Akio Nagasaka, Takafumi Miyatake. Feature Extraction of Finger Vein Patterns Based on Repeated Line Tracking and Its Application to Personal Identification, HITACHI, Ltd. 1-280, Machine Vision and Applications (2004)
[2] Naoto Miura, Akio Nagasaka, and Takafumi Miyatake. Feature Extraction of Finger Vein Patterns Based on Iterative Line Tracking and Its Application to Personal Identification, System and Computers in Japan, Vol. 35, No. 7, 2004
[3]Yuhan Ding, Dayan Zhuang, and Kejun Wang. A Study of Hand Vein Recognition Method, Department of Automation, Harbin Engineering University. In: Proceeding of the IEEE, International Conference on Mechatronics & Automation, July 2005
[4]Kejun Wang, Yan Zhang, Zhi Yuan, and Dayan Zhuang. Hand Vein Recognition Based on Multi Supplemental Features of Multi-Classifier Fusion Detection, Department of Automation, Harbin Engineering University. In: Proceeding of he 2006 IEEE, International Conference on Mechatronics and Automation in Luoyang - China, June 2006
[5] Wang Lingyu and Graham Leedham. Near and Far Infrared Imaging for Vein Pattern Biometrics, School of Computer Engineering - Nanyang Technological University Singapore; UNSW Asia Singapore. In: Proceeding of the IEEE, International Conference on Video and Signal Based Surveillance (AVSS’06), 2006
[6] Hitachi Review Vol. 53, No. 2. Door-access-control System Based on Finger Vein Authentication, 2004
[7] Kresimir Delac, Mislav Grgic. A Survey of Biometric Recognition Methods, HT Croatian Telecom – Carrier Services Department; University of Zagreb,

CROATIA. In: 46th International Symposium Electronics in Marine, ELMAR-2004
[8] Sangkyun Im and Hwansoo Choi. A Filter Bank Algorithm for Hand Vascular Pattern Biometrics, Tech Sphere Co., Ltd; Department of Information Control Engineering, Myong-Ji University. In: Seventh International Conference on Control, Automation, Robotics and Vision (ICARCV’02), Dec 2002
[9] Prof. Dr-Ing. Jana Dittman. Multimedia and Security.
[10] Junichi Hashimoto. Finger Vein Authentication Technology and Its Future. Information & Telecommunication System Group, Hitachi, Ltd. In: Symposium on VLSI Circuits Digest of Technical Papers, 2006
[11] National Science and Technology Council (NSTC), Committee on Technology, Committee on Homeland and National Security, Subcommittee on Biometrics. Vascular Pattern Recognition, 2006
[12]Brice O. and Stephane R. Biometrics: Palm Vein Authentication, Fujitsu Laboratories Limited, 2005
[13]T. Yanagawa, S. Aoki, and T. Ohyama. Human Finger Vein Images are Diverse and Its Patterns are Useful for Personal Identification, MHF Preprint Series, Faculty of Mathematics, Kyushu University, Fukuoka – JAPAN, April 2007
[14] Slobodan Ribaric and Ivan Fratric. An Online Biometric Authentications System Based on Eigenfingers and Finger Geometry, Faculty of Electrical Engineering and Computing, University of Zagreb, Croatia.

QR CODE