隨著科技日新月異的發展、學者們精益求精的精神以及電腦視覺專家們互相 切磋琢磨的風氣，使得電腦視覺領域的發展突飛猛進，也得以讓生活變得更方便 美好。以前傳統的身分認證方法僅侷限於晶片卡，之後漸漸地演變成指紋辨識、
虹膜辨識以及人臉辨識。本論文提出一套兩階段的人臉辨識技術，第一階段是使 用基於人臉區塊位置而給予權重的 WLVP 模組來進行相似候選人篩選，而第二 階段是使用以特徵點為基礎的雙向辨識演算法，使用特徵點匹配並搭配特徵點之 間的幾何結構關係來進行最後的辨識運算。根據實驗結果可得知，本論文的辨識 技術在人臉辨識的應用上，具有相當高準確性的辨識率。
另外，因為當特徵點數目較多時，特徵點匹配需要較長的運算時間，可能會 讓使用者必須稍微等待而產生不便。雖然降低特徵點的數量，可以減少運算時間，
但是任意的刪減特徵點可能會影響到辨識效果，所以應該如何篩選出穩定且具鑑 別性的特徵點以縮短辨識的運算時間，甚至能進一步地提升辨識率，將是未來需 要繼續努力的重要課題。
45
參考文獻
[1] M. Turk and A. Pentland, “Eigenfaces for Recognition”, Journal of Cognitive Neuroscience, Vol. 3, No. 1, pp. 71-86, 1991.
[2] V. Belhumeur, J. Hespanha, and D. “Kriegman. Eigenfaces vs. Fisherfaces:
Recognition using Class Specific Linear Projection”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, pp. 711-720, July 1997.
[3] D. Cai, X. He, J. Han, and H. Zhang, “Orthogonal Laplacianfaces for Face Recognition”, IEEE Transactions on Image Processing, Vol. 15, No. 11, pp.
3608-3614, 2006.
[4] J. Shawe-Taylor and N. Cristianini, “Kernel Methods for Pattern Analysis”, Cambridge University Press, 2004.
[5] W. Liu, J. Principe and S. Haykin, “Kernel Adaptive Filtering: A Comprehensive Introduction”, Wiley, 2010.
[6] Hong Chang & Dit-Yan Yeung, “Robust Locally Linear Embedding”, Technical Report HKUST-CS05-12, 2005.
[7] G. Baudat and F. Anouar, “Generalized Discriminant Analysis using a Kernel Approach”, Neural Comput., Vol. 12, pp. 2385–2404, 2000.
[8] T. Ojala, M. Pietikäinen and T. Mäenpää, “Multi-Resolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns”, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 24, pp. 971-987, 2002.
[9] S. Murala, R. P. Maheshwari, and R. Balasubramanian, “Local Tetra Patterns: A New Feature Descriptor for Content-Based Image Retrieval”, IEEE Trans.
Image Process., Vol. 21, No. 5, pp. 2874-2886, May 2012.
[10] Fan, K., Hung, T., "A Novel Local Pattern Descriptor—Local Vector Pattern in
46
High-Order Derivative Space for Face Recognition", IEEE Trans. Image Process., Vol. 23, No. 7, pp. 2877-2891, July 2014.
[11] Ngoc-Son Vu, Alice Caplier, "Face Recognition with Patterns of Oriented Edge Magnitudes", In: K. Daniilidis, P. Maragos, N. Paragios (Eds.): ECCV 2010, Part I, LNCS 6311, pp. 313–326 (2010).
[12] David G. Lowe, “Distinctive Image Features from Scale-Invariant Keypoints”, International Journal of Computer Vision, Vol. 60, No. 2, pp. 91-110, 2004.
[13] H. Bay, T. Tuytelaars, and L. V. Gool, “SURF: Speeded Up Robust Features”, in Proc. of the 9th European Conference on Computer Vision (ECCV’06), ser.
Lecture Notes in Computer Science, A. Leonardis., 2006.
[14] Brown, M.; Lowe, David G. (2003)., "Recognizing Panoramas", Proceedings of the ninth IEEE International Conference on Computer Vision. 2. pp. 1218–1225.
doi:10.1109/ICCV.2003.1238630, 2003
[15] Y.S. Huang, H.Y. Cheng, P.F. Cheng, C.Y. Tang, “Face Detection with High Precision Based on Radial-Symmetry Transform and Eye-Pair Checking”, In IEEE Conference on Advanced Video and Signal Based Surveillance, pp. 62, 2006.
[16] P.A. Viola, M.J. Jones, “Rapid Object Detection Using a Boosted Cascade of Simple Features”, CVPR, Issue 1, pp. 511–518., 2001.
[17] A. Neubeck, L. Van Gool, “Efficient Non-Maximum Suppression”, In: ICPR, 2006.
[18] M. Brown, D. Lowe, “Invariant Features from Feature Point Groups”, In: BMVC, 2002.
[19] T. Ahonen, A. Hadid, and M. Pietik¨ainen, “Face Description with Local Binary Patterns: Application to Face Recognition”, IEEE Trans. Pattern Anal. Mach.
Intell., Vol. 28, No. 12, pp. 2037-2041, Dec. 2006.
47
[20] S. Bengio, F. Bimbot, J. Mari´ethoz, V. Popovici, F. Por´ee, E. Bailly- Bailli`ere, G. Matas, and B. Ruiz, “Experimental Protocol on the BANCA Database”, IDIAP-RR 05, IDIAP, 2002.
[21] P.J. Phillips, H. Wechsler, J.S. Huang, and P.J. Rauss, “The FERET Database and Evaluation Procedure for Face-Recognition Algorithms”, Image and Vision Computing, Vol. 16, No. 5, pp. 295-306, 1998.
[22] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman,“Eigenfaces vs. Fisherfaces:
Recognition using Class Specific Linear Projection,” IEEE Trans. Pattern Anal.
Mach. Intell., Vol. 19, No. 7, pp. 711-720, Jul. 1997.
[23] Wiskott, L. , Fellous, J.-M. ,Kuiger, N. ,von der Malsburg, C., "Face Recognition by Elastic Bunch Graph Matching", IEEE Trans. Pattern Anal. Mach. Intell., Vol. 19 , No. 7, pp. 775-779, Jul. 1997.
[24] T. Ojala, M. Pietikäinen, and T. Mäenpää, “Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns,” IEEE Trans. Pattern Anal. Mach. Intell., Vol. 24, No. 7, pp. 971–987, Jul. 2002.
[25] T. Ahonen, A. Hadid, and M. Pietikäinen, “Face Description with Local Binary Patterns: Application to Face Recognition,” IEEE Trans. Pattern Anal. Mach.
Intell., Vol. 28, No. 12, pp. 2037–2041, Dec. 2006.
[26] B. Zhang, Y. Gao, S. Zhao, and J. Liu, “Local Derivative Pattern Versus Local Binary Pattern: Face Recognition with Higher-Order Local Pattern Descriptor,”
IEEE Trans. Image Process., Vol. 19, No. 2, pp. 533-544, Feb. 2010..
[27] S. Murala, R. P. Maheshwari, and R. Balasubramanian, “Local Tetra Patterns: A New Feature Descriptor for Content-Based Image Retrieval,” IEEE Trans. Image Process., Vol. 21, No. 5, pp. 2874–2886, May 2012.
[28] B. Zhang, Y. Gao, S. Zhao, and J. Liu, “Local Derivative Pattern Versus Local Binary Pattern: Face Recognition with Higher-Order Local Pattern Descriptor,”
48
IEEE Trans. Image Process., Vol. 19, No. 2, pp. 533-544, Feb. 2010.
[29] T. Ojala, M. Pietika¨inen, and T. Ma¨enpa¨a¨, “Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 24, No. 7, pp. 971-987, July 2002.
49
附錄 A
𝑆2(∙,∙)轉換函式定義為
else.
, 0
0 ) ( )
) ( (
) if (
, 1
)) ( ),
( ), ( ),
( (
, , 45 ,
, ,
, 45
, 45 ,
, , 45 ,
, 2
R p D R
p D c
D c D
c D c
D R p D R
p D
G V
G G V
V
G V
G V
G V G V
G V S
其中，比較的空間轉換(CST)式子如下
0 ) ( )
) ( (
) (
, , 45 ,
, ,
,
45
R p D R
p D c
D c
D
V G V G
G V
G V
在此以圖 A-1 來解釋 CST 的設計原理。
圖 A-1、鄰居𝐺7,𝑅的β方向之 LVP 編碼示意圖
圖 A-1 中紅圈表示參考點𝐺𝑐，黃圈表示𝐺𝑐之鄰居點𝐺7,𝑅，而每個格子內的數字代 表該像素點的亮度值。當 D 為 1 而β為 0 時，𝐺7,𝑅於(β + 45°)方向之鄰近像素點 的值為 9，而 𝐺7,𝑅於(β)方向之鄰近像素點的值為 8，此二數值相減則得到
) (
7,,
45 D
G
RV
值為 1。使用相同的運算，則可以得到V
45,D( G
c)
、V
,D( G
c)
和)
(
7,,D