臺灣博碩士論文加值系統

本論文提出量化模糊集中二元圖樣(QFCBP)的人臉辨識演算法，並與其他二元圖樣演算法做運算速度與辨識率的比較。文獻上有很多不同的二元圖樣演算法，像是區域二元圖樣演算法(LBP)、比區域二元圖樣演算法有更好辨識率的模糊區域二元圖樣演算法(FLBP)、能減少運算量的集中二元圖樣演算法(CBP)以及模糊集中二元圖樣(FCBP)。我們採用兩個人臉資料庫來驗證所提之量化模糊集中二元圖樣演算法，實驗顯示以權重卡氏平方距離 (WCSD)作人臉辨識評量，所提之演算法能達到高運算速度與高辨識率。

This thesis proposes a quantized fuzzy centralized binary pattern (QFCBP) method for face recognition. The proposed QFCBP method is demonstrated by comparison with other binary pattern methods in term of computational speed and recognition accuracy. In the literature, there are many binary pattern methods for face recognition, including local binary pattern (LBP) which is invariant to gray-scale changes of the image, fuzzy local binary pattern (FLBP) which has better recognition accuracy than LBP, centralized binary pattern (CBP) which reduces the histogram dimensionality and considers the center pixel effect, and fuzzy centralized binary patterns (FCBP). Experiments on two facial databases show that our QFCBP method can achieve high computational speed and high recognition accuracy when the weighted Chi square distance (WCSD) is used for similarity measurement of face recognition.

Chinese Abstract i
English Abstract ii
Tables of Contents iii
List of Tables v
List of Figures vi
CHAPTER 1 Introduction 1
1.1 Background 1
1.2 Organization of the Thesis 2
CHAPTER 2 Review of Binary Pattern Methods for face recognition 3
2.1 Local Binary Pattern 3
2.2 Centralized Binary Pattern 5
2.3 Fuzzy Local Binary Pattern 6
2.4 Fuzzy Centralized Binary Pattern 8
2.5 Quantized Fuzzy Local Binary Pattern 10
CHAPTER 3 Quantized Fuzzy Centralized Binary Pattern 12
3.1 Introduction 12
3.2 Proposed QFCBP Method 13
3.3 Weighted Chi Square Distance 14
CHAPTER 4 Experimental Results 16
4.1 Introduction 16
4.2 JAFFE database 16
4.3 Cohn-Kanade database 19
CHAPTER 5 Conclusion 23
5.1 Conclusion 23
References 24

[1] M. Turk and A. Pentland, “Eigenfaces for recognition constraints,” Journal of Cognitive Neuroscience, vol. 3, pp. 71-86, 1991.
[2] K. Etemad, and R. Chellappa, “Discriminant analysis for recognition of human face images,” J. of Opt. Soc. Am. ,vol.14, pp. 1724-1733, 1997.
[3] M. S. Bartlett, J. R. Movellan and T. J. Sejnowski, “Face recognition by independent component analysis,” IEEE Trans. Neural Netw. , vol.13, pp. 1450-1464, 2002.
[4] T. Ojala, M. Pierikainen, and M. Maenpaa, “Multiresolution grayscale and rotation invariant texture classification with local binary pattern,” IEEE trans. Pattern Anal. Mach. Intell., vol.24, pp. 971-987, 2002.
[5] X. Fu and W. Wei, “Centralized Binary Patterns Embedded with Image Euclidean Distance for Facial Expression Recognition, ” International Conference on Natural Computation, vol.4, pp. 115-119,2008.
[6] D. K. Iakovidis, E. G. Keramidas, and D. Maroulis, “Fuzzy local binary patterns for ultrasound texture characterization, ” in Image Analysis and Recognition, ed: Springer, pp. 750-759, 2008..
[7] M. R. Mohammadi and E. Fatemizadeh, “Fuzzy Local Binary Patterns: A Comparison between Min-Max and Dot-Sum Operators in the Application of Facial Expression Recognition,” IEEE Iranian Conference on Machine Vision and Image Processing (MVIP), pp. 315-319, 2013.
[8] T. Ahonen, A. Hadid and M. Pietikainen, “Face Description with Local Binary Patterns: Application to Face Recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.28, pp. 2037-2041, Oct. 2006.
[9] C. C. Chiu, “Fuzzy Centralized Binary Patterns Embedded with Histogram for Face Recognition,” Master thesis of Tatung University, 2016.
[10] J. Ren, X. Jiang and J. Yuan, “Quantized Fuzzy LBP for Face Recognition,” IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp 1503 - 1507, April. 2015.