臺灣博碩士論文加值系統

　　隨著科技進步，為了提供給使用者更多的便利和可靠性，所以漸漸開始使用人臉辨識、語音辨識、指紋辨識、掌紋辨識和虹膜辨識等相關的生物認證技術。
　　在計算機視覺中，圖像識別具有十分重要的意義，而其中不變矩的圖像平移、縮放、旋轉均保持不變是圖像識別的重要方法之一，1961年，Hu提出了七個不變矩用於圖像識別，利用不變矩進行識別獲得了廣泛的應用。在論文中的虹膜辨識系統，使用四元數結合不變矩，並替換核心函數去討論其不變性。
　　實驗結果顯示，改良後的四元數不變矩能夠應用於虹膜辨識系統，雖然初步的實驗結果不理想，僅跟傳統不變矩的辨識率相近，但是比起傳統的方法，少了一些影像處理步驟，推論加強影像前處理模組或分類辨識模組替換，可以提升辨識率。

With the development of science and technology, the technology provide users with more convenience and reliability. The biometric technology such as the face recognition, voice recognition, fingerprint recognition, palm print recognition and iris recognition are widely used.
Pattern recognition is important in computer vision. One of the important method is that the translation, scaling, rotation remain unchanged in moments invariant. In 1961, Hu proposed seven moments invariant for Pattern recognition and have been used widely.
In the thesis, the iris recognition system uses the quaternion with moments invariant and discuss with the invariant of replacing the kernel function.
The experimental results reveal that the improved quaternion moments invariant can be applied to iris recognition system. Although the preliminary experimental results are not satisfactory. However, When the improved quaternion moments compared to traditional methods, the reduce the a pretreatment steps think that strengthening the image preprocessing module or classification identification module replacement can enhance the recognition rate.

誌謝 I
摘要 II
ABSTRACT III
目錄 IV
圖目錄 VI
表目錄 VII
第一章　緒論 1
1.1研究動機 1
1.2研究目標與方向 2
1.3論文架構 3
第二章　應用於虹膜之影像前處理技術 5
2.1色彩模型 5
2.1.1 RGB色彩模型 5
2.1.2 HSV色彩模型 6
2.2影像強化 7
2.2.1灰階轉換 7
2.3圓與曲線偵測 7
2.3.1圓偵測 8
2.3.2曲線偵測 9
第三章　特徵萃取文獻回顧 11
3.1　Hu Moment 12
第四章　四元數不變矩 17
4.1四元數 17
4.1.1矩陣表示 17
4.2四元數矩 18
4.2.1四元數矩的定義 18
4.2.2四元數的仿射不變 19
4.3改良之四元數不變矩 21
第五章　基於四元數不變矩的彩色影像分析 23
5.1系統架構 23
5.2前處理模組 24
5.2.1虹膜定位 24
5.2.2虹膜分割與正規化 24
5.3特徵萃取模組 25
5.4分類辨識模組 25
5.4.1歐式距離 25
第六章　實驗結果 26
6.1實驗環境 26
6.2實驗說明 26
6.3系統效能評估 27
6.4實驗結果 28
第七章　結論 32
參考文獻 33

[1] M.K. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory IT-8 (1962) 179–187.
[2] J.L. Coatrieux, “Moment-Based Approaches in Imaging. 1. Basic Features [A Look At ...] ,” IEEE, Engineering in Medicine and Biology Magazine, Vol. 26, Issue 5, October 2007
[3] J.L. Coatrieux, “Moment-based approaches in imaging part 2: invariance [A Look at . . .] ,” IEEE, Engineering in Medicine and Biology Magazine, Vol. 27, Issue 1, February 2008
[4] J.L. Coatrieux, “Moment-based approaches in imaging part 3: computational considerations [A Look at...] ,” IEEE, Engineering in Medicine and Biology Magazine, Vol. 27, Issue 3, June 2008
[5] S.C. Pei and C.M. Cheng, Color Image Processing by Using Binary Quaternion Moment Preserving Thresholding Technique,” IEEE, Transactions on Image Processing, Vol. 8, No. 5, May 1999.
[6] B.J. Chen and H.Z. Shu, “Quaternion Zernike moments and their invariants for color image analysis and object recognition,”IEEE, Signal Processing, pp. 308–318, 2012.
[7] A. Khotanzad and Y.H. Hong, “Invariant image recognition by Zernike moments,” IEEE Trans. Pattern Anal. Mach. Intell., 1990.
[8] 朱明, 孫繼剛和郭立強, 彩色圖像的四元數不變矩, 中國科學院長春光學精密機械與物理研究所, 2011.
[9] 曹駿, 利用序列虹膜影像進行辨識與抗偽造之研究, 國立暨南國際大學, 碩士論文, 民國94年.
[10] 黃仁宏, 基於三維灰階伴隨矩陣之虹膜辨識, 國立暨南國際大學, 碩士論文, 民國97年.
[11] 蔣宜璋, 虹膜辨識系統實作, 國立暨南國際大學, 碩士論文, 民國99年.
[12] W.R. Hamilton, “Elements of Quaternions,” Longmans, Green, London, 1866.
[13] “http://caterpillar.onlyfun.net/Gossip/ComputerGraphics/QuaternionsRotate.htm,” 四元數與旋轉.
[14] J.G. Daugman, “How iris recognition works,” IEEE Trans. On Circuits and Systems for Video Technology, Vol. 14, No. 1, pp. 21-30, Jan. 2004.
[15] J.G. Daugman, “High confidence visual recognition of persons by a test of statistical independence,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 15, No. 11, pp. 1148-1161, Nov. 1993.
[16] R.P. Wildes, “Iris recognition: an emerging biometric technology,” Proc. of the IEEE, Vol. 85, No. 9, pp. 1348-1363, Sep. 1997.
[17] 核心函數定義參考來源http://blog.sciencenet.cn/blog-577247-590122.html