傳統的身份辨識工具，如鑰匙、磁卡與密碼，常有被人複製與破解的風險。而沒帶鑰匙、遺失磁卡或忘記密碼的情境，在吾人周遭，屢見不鮮，實造成生活上與工作上的諸多不便和不安。近年來由於科技的發達，生物特徵因具唯一性、不變性以及難仿造性，常被用來作為相對方便與安全的身份認證工具。一般用來設計辨識系統的生物特徵，有臉部、虹膜、視網膜、掌紋、指紋與聲紋，等諸多選項。然而，首先根據 Leonard Flom 與 Aran Safir 的統計研究，不同人出現相同虹膜的機率約為 1/〖10〗^78。以目前全球約 72.5億 (7.25×〖10〗^9) 總人口來看，出現相同虹膜的機率，實在微乎其微。
其次，「虹膜」是由眼球上的纖維與肌肉所組成，為眼球上最豐富的紋理。虹膜的紋理，於胎兒出生後，半年至一年半間，已幾乎定型。最後，仿造虹膜的工藝難度，遠超過人臉與指紋。因此，從唯一性、不變性與難仿造性的角度來看，虹膜當作生物辨識的基礎，是相當穩當的。在本論文中，吾人設計與實作了一套虹膜辨識系統，用以驗證系統的可行性，並探討其設計理念，與評量其系統效能。
經由鏡頭原理的研究，吾人改良了市售的紅外線攝影機，獲取了清晰的虹膜影像。系統先進行瞳孔偵測，將虹膜定位，最後透過局部二元化與傅立葉轉換的數理運算，萃取虹膜中具有方向性的特徵，來做未知身份的認證比對。
本系統在Intel Core i5筆電與Windows 8的作業系統環境下，對實驗室七人，每人取16張左眼虹膜影像，分四天進行拍攝，每天拍攝四張，第一天的四張虹膜，作為訓練資料，剩下的12張，作為測試影像，經各特徵參數加權比對後，正確率可達100%。另外，針對 IITD，CASIA-Interval 與 CASIA-Iris V1 虹膜資料庫，所分別提供的224，249與108人之虹膜，辨識率亦可達87.08%，89.05%和76.54%。

Conventional identification tools, such as keys, magnetic cards and secret codes, can be easily duplicated and cracked. Missing keys, losing cards and forgetting passwords are quite usual that result in inconvenience and insecurity in our daily life and work. In the recent years, biometrics technology is applied in the authentication industry due to its unique, stable and non-counterfeitable features. Face, iris, retina, palm print, fingerprint and voiceprint are the common biometric features. However, first of all, according to the statistical analysis of Leonard Flom & Aran Safir, the probability of occurrence that different people have the same iris is about 1/〖10〗^78. The chance of two people in our planet have the same iris is extremely trifling based on the fact that the current global population is approximate 7.25 billion (7.25×〖10〗^9).
Secondly, "Iris" is composed of fibrous tunic and muscle tissue in the eyeball, and exhibits significantly rich texture. Furthermore, the iris pattern of a person is stable and unchanged after six to eighteen months of his birth. Finally, the degree of difficulty of making a fake iris is far beyond that of a face image or a fingerprint from the technology point of view. It is concluded that iris is a surpassingly reliable biometric feature due to its unique, stable and non-counterfeitable feature. In this thesis, an iris recognition system is designed and implemented to justify the feasibility of this technology. The design concept and system performance are also introduced and evaluated.
An infrared camera for the iris image capture is made in our research by modifying a commodity video camera through a study on optical lens theory. Clear pictures of iris images are first acquired. Pupil detection and iris pattern segmentation are then applied. Finally, mathematical operations of local binary patterns and Fourier transform are utilized to extract the directional features of the iris to recognize the unknown user.
Under the Intel Core i5 notebook and Windows 8 operating system environment, a correct recognition rate of 100% can be obtained by the weighted feature strategy for a 7 person left-eye iris database, each with 16 images recorded in 4 days. The 4 images in the first day are used for training, and the other 12 images are for testing. In addition, the system is also tested on the 224 users of IITD iris database, the 249 users of CASIA-Interval iris database and the 108 users of CASIA V1 iris database, correct rates of 87.08%, 89.05% and 76.54% can be achieved respectively.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖目錄 vi
表目錄 ix
第一章 緒論 1
1.1研究動機 1
1.2研究目的與方法 2
1.3論文章節概要 2
第二章 虹膜特徵介紹 3
2.1虹膜學 3
2.2虹膜細部介紹 4
2.3虹膜紋理特徵 5
第三章 虹膜拍攝方法與拍攝設備介紹 8
3.1凸透鏡 8
3.1.1凸透鏡介紹 8
3.1.2微距原理 12
3.2拍攝照明 16
3.3 拍攝之攝影機 17
第四章 辨識系統之流程與架構 20
4.1辨識系統之流程 20
4.2影像前置處裡 21
4.2.1影像二值化 22
4.2.4 正規化 35
4.2.5 極座標轉換 36
4.3 虹膜特徵萃取 37
第五章 辨識系統之實作成果與效能評析 42
5.1 軟硬體設備與開發環境 42
5.2 影像模型建立 43
5.3 權重分析比對 44
5.4 系統辨識效能 45
第六章 結論與未來展望 50
參考文獻 51

[1] L.Flom and A.Safir, “Iris Recognition System,” U.S.Patent, no.4641349, Feb 1987.
[2] R. Wildes, “Iris Recognition: An Emerging Biometric Technology,” Proceedings of the IEEE, Vol. 85, pp. 1348-1363, Sep 1997.
[3] John G. Daugman, “High Confidence Recognition of Persons by Iris Patterns,” IEEE 35th International Carnahan conference on Security Technology, 254~263, 2001.
[4] John G. Daugman, “High Confidence Visual Recognition of Persons by a Test of Statical Independence,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.15, no.11, pp.1148~1161, Nov 1933.
[5] Timo Ojala, Matti Pietikainen and T. Maenpaa, “Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, No. 7, pp. 971-987, 2002
[6] Qi-Chuan Tien et al., “Fast Algorithm and Application of Hough Transform in Iris Segmentation,” Proc. Of the 3th International Conference on Machine Learning and Cybernatics, Shanghai, Aug 2004.
[7] R. S. Feris, T. E. Campos, and R. M. Cesar, “Detection and Tracking of Facial Features in Video Sequences,” in Mexican International Conference on Artificial Intelligence, vol. 1793, pp. 129-137, Apr 2000.
[8] Sheeba Jeya Sophia S. and Veluchamy S., “Security System Based on Iris Recognition,” Research Journal of Engineering Sciences, Vol. 2(3), 16-21, Mar 2013.
[9] R. L. Hsu, M. A. Mottaleb, and A. K. Jain, “Face Detection in Color Images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, pp. 696-706, May 2002.
[10] Institute of Automation, Chinese Academy of Sciences, CASIA-Iris V1, http://biometrics.idealtest.org/
[11] Institute of Automation, Chinese Academy of Sciences, CASIA Iris Image Database, http://www.sinobiometrics.com/
[12] IIT Delhi Iris Database (Version 1.0) : http://www4.comp.polyu.edu.hk/~csajaykr/IITD/Database_Iris.htm
[13] http://www.dls.ym.edu.tw/neuroscience/bigeye_c.html
[14] 虹膜學，取自台灣國際虹膜學會http://www.iris-tw.com/
[15] 劉靜怡、陳顯武、周桂田、廖福特、吳豪人（民95），運用生物特徵辨識身分制度之比較研究，臺北市：行政院研究發展考核委員會。
[16] Gonzalez Woods（民98），數位影像處理（繆紹綱譯），臺北縣：台灣培生教育。
[17] 鍾國亮（民101），影像處理與電腦視覺，臺北市：東華。
[18] 莊英杰，追瞳系統之研發於身障者之人機介面應用，國立中央大學資訊工程研究所碩士論文，民國九十三年。
[19] 紅外線，取自遠紅外線醫療網
http://www.far-infrared.info/subject/FarInfraRayKnowledge.aspx?item=41
[20] 楊長璟，傅立葉轉換於虹膜辮識之應用，台灣科技大學電機工程研究所碩士論文，2001。
[21] 林鳳軒，呂愛玉，虹彩診察術，美加生命能量虹彩研究協會，2002。