現今人臉辨識系統從訓練資料使用有限的特徵或模組，而不是直接使用所有的資料來表示或分類輸入圖片的趨勢越來越盛行。稀疏表示(SRC)透過l1最小化將多張訓練圖片線性組合成一張測試輸入圖，且已證實其在人臉辨識系統上表現相當好。然而，預先使用一個特徵描述器做前處理能夠改善辨識率，因為相較於把原始人臉圖片的像素做SRC，我們使用SRC在一個特徵描述的龐大資料集上。此研究顯示結合兩個現今很成功的特徵擷取演算法，賈伯濾波器與局部二元模式(LBP)，能有更好的效果。此外，我們將不會在每次有新的測試圖片輸入時，從頭開始再做一次l1最小化來解最佳化，我們反而是更新上一次的解，以它做為開頭與基礎並計算是要加上或是減去矩陣中的元素，然後隨著相較之下沒那麼冗長的步驟跟著更新並得到最終解。我們在兩個指標人臉資料庫Extended Yale B與AR上測試提出的方法，並展示此方法在亮度變化與遮蔽下的人臉辨是效果相當不錯。

It has become more popular to use a limited subset of features from the training data, rather than directly use the data for representing an input image. Sparse Representation based classification (SRC) represents the input test image as a linear combination of the training samples via l1-minimization, and proves to work well on face recognition. However, the prior use of a discriminative descriptor improves the recognition rate because instead of applying SRC straight to raw face image pixels, we apply it on a large dictionary of keypoint descriptors. We demonstrate that combining two of the most successful feature extraction algorithms, Gabor Filter and Local Binary Pattern (LBP) yields a better performance overall. Moreover, we won’t do l1-minimization afresh every time when the system gets a new test image, instead we update the previous solution, which is used as a start-up, and subsequently add or remove measurements in the system, then update the solution accordingly in a sequence of steps that are cheaper in terms of time complexity. Experiments on popular face databases, Extended Yale B and AR, show that the proposed method handles face recognition under illumination changes and occlusions quite well.

[1] M. Turk and A.P. Pentland, “Face Recognition Using Eigenfaces,” IEEE Conf. Computer Vision and Pattern Recognition, 1991.
[2] P.N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, “Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection,” IEEE Transaction on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, July 1997.
[3] X. He and P. Niyogi, “Locality Preserving Projections,” Proc. Conf. Advances in Neural Information Processing Systems, 2003.
[4] T. Ojala, M. Pietikäinen, and D. Harwood, “A Comparative Study of Texture Measures with classification Based on Feature Distributions,” Pattern Recognition, vol. 29, no.1, pp. 51-59, 1996
[5] X. Tan and B. Triggs, “Enhanced Local Texture Feature Sets for Face Recognition under Difficult Lighting Conditions,” Proc. IEEE Int’l Workshop Analysis and Modeling of Faces and Gestures, 2007
[6] J. Wright, A.Y. Yang, A. Ganesh, S.S. Sastry, and Y. Ma, “Robust Face Recognition via Sparse Representation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 31, no. 2, pp. 210-227, Feb. 2009.
[7] L. Wiskott, J.-M. Fellous, N. Kruger, and C. von der Malsburg, “Face Recognition by Elastic Bunch Graph Matching,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 775-779, July 1997.
[8] C. Liu and H. Wechsler, “Gabor Feature Based Classification Using the Enhanced Fisher Linear Discriminant Model for Face Recognition,” IEEE Trans. Image Processing, vol. 11, no. 4, pp. 467-476, Apr. 2002.
[9] E. Candes and J. Romber, “l1-Magic: Recovery of Sparse Signals vis Convex Programming,” http://www.acm.caltech.edu/l1magic/, 2005.
[10] S. Boyd and L. Vandenberghe, Convex Optimization. Cambridge University. Press, 2004.
[11] X. He, S. Yan, Y. Hu, P. Niyogi, and H.-J. Zhang, “Face Recognition Using Laplacianfaces,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 3, March 2005.
[12] V. Vapnik, The Nature of Statistical Learning Theory. Springer, 2000.
[13] T. Cover, “Geometrical and Statistical Properties of Systems of Linear Inequalities with Applications in Pattern Recognition,” IEEE Trans. Electronic Computers. vol. 14, no. 3, pp. 326-334, 1965.
[14] R. Duda, P. Hart, and D. Stock, Pattern Classification, second ed. John Wiley & Sons, 2001.
[15] J. Ho, M. Yang, J. Lim, K. Lee, and D. Kriegman, “Clustering Appearances of Objects under Varying Illumination Conditions,” Proc, IEEE Int’l Conf. Computer Vision and Pattern Recognition, pp. 11-18, 2003.
[16] M. Lades, J. C. Vorbruggen, J. Buhmann, J. Lange, C. von der Malsburg, R. P. Wurtz, and W. Konen, “Distortion Invariant Object Recognition in the Dynamic Link Architecture,” IEEE Trans. On Computers, vol. 42, no. 3, March 1993.
[17] R. Basri and D. Jacobs, “Lambertian Reflection and Linear Subspaces,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 25, no. 3, pp. 218-233, Mar. 2003.
[18] D. Donoho, “For Most Large Underdetermined Systems of Linear Equations the Minimal l1-Norm Solution Is Also the Sparsest Solution,” Comm. Pure and Applied Math., vol. 59, no. 6, pp, 797-829, 2006.
[19] E. Candes and T. Tao, “Near-Optimal Signal Recovery from Random Projections: Universal Encoding Strategies?” IEEE Trans. Information Theory, vol. 52, no 12, pp. 5406-5425, 2006.
[20] E, Candes, J. Romberg, and T. Tao, “Stable Signal Recovery from Incomplete and Inaccurate Measurements,” Comm Pure and Applied Math., vol. 59, no. 8, pp. 1207-1223, 2006.
[21] M. S. Asif and J. Romber, “Dynamic Updating for l1 Minimization,” IEEE Journal of Selected Topics in Signal Processing, vol. 4, no. 2, pp. 421-434, Apr. 2010.
[22] A. Martinez and R. Benavente, “The AR Face Database,” CVC Technical Report 24, 1998.
[23] A. Georghiades, P. Belhumeur, and D. Kriegman, “From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 23, no. 6, pp. 643-660, June 2001.