||Image-to-Class Warping with Threshold-LBP for Face Recognition
||Department of Electrical Engineering
Threshold Local Binary Pattern
隨著科技的日新月異，人臉辨識系統已經與生活密不可分。人臉辨識系統被廣泛應用在許多領域中，像是相機對焦、安全系統和信用卡認證等。但是人臉辨識系統往往會受到一些無法控制的環境因素導致辨識率下降，像是光照變化、姿勢、表情與遮蔽物。為了解決這些問題我們會從訓練資料中擷取出人臉影像的特徵再透過分類器分類，然後將測試影像與資料庫影像進行比對。在這篇論文中，我們將原始影像透過閾值局部二進模式(TLPB)表示，再將影像分成許多小區塊。不需經過訓練階段直接以塊狀匹配的方式計算歐幾里得距離。塊狀匹配並非計算影像至影像距離而是影像至類別距離。我們在Aleix Martinez and Robert Benavente (AR) 所建立的人臉資料庫上測試提出的方法，並展示此方法在遮蔽下的人臉辨識有不錯的辨識效果。
With the rapidly-changing technology, face recognition system has been inseparable from our life. It is widely used in many areas, such as camera focus, security systems and credit card authentication. But the face recognition system is often subject to some uncontrollable environmental factors leading to decreased identification rate, such as variable illumination, postures, expressions and occlusions. In order to solve these problems, we will extract the features of face images from the training image sets and classify the test image by the classifier. Then compare the test image to the database images. In this paper, we represent the original image with the threshold local binary pattern (TLPB), and then the image is divided into many small patches called sub-patches. Without the training phase, we directly calculate the overall cost by conducting image-to-class warping. Image-to-class warping is similar to the patch-matching method. Patch-matching does not calculate the image-to-image distance but the image-to-class distance. We test the proposed method on the Aleix Martinez and Robert Benavente face database and show that the method has a good performance in the situation that the occlusions occur in either probe image set or gallery image set.
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Face Recognition 2
1.2.1 Common FR system 2
1.2.2 Occlusions handling 3
1.3 Features Extraction 3
1.4 Classification 5
1.5 Thesis Organization 5
Chapter 2 Related Work 7
2.1 Features Extraction 7
2.1.1 Local Binary Pattern 7
2.1.2 Threshold Local Binary Pattern 9
2.2 Classification 11
2.2.1 Euclidean Distance 11
2.2.2 K-Nearest Neighbors (KNN) 11
Chapter 3 Proposed Method 14
3.1 Framework 14
3.2 Image Representation 16
3.2.1 Threshold Local Binary Pattern Descriptor 16
3.2.2 Patch Sequence 17
3.2.3 Difference Patch Sequence 17
3.3 Image-to-Class Warping 19
3.4 Classification 21
Chapter 4 Experimental Results 22
4.1 Face Database 22
4.2 Face Identification with Different Conditions 23
4.2.1 Un-occluded vs. Un-occluded 24
4.2.2 Occluded vs. Un-occluded 25
4.2.3 Un-occluded vs. Occluded 28
4.3 Compared with Image-to-Image Distance 29
4.4 The Effect of Different Patch Size 30
Chapter 5 Conclusions 32
 M. Turk and A.P. Pentland, “Face Recognition Using Eigenfaces”, IEEE Conf. Computer Vision and Pattern Recognition, 1991.
 P. N. Belhumeur, J. P. Hespanha and D. J. 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, 1997.
 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.
 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.
 Jun Meng, Yumao Gao, Xiukun Wang, Tsauyoung Lin and Jianying Zhang, “Face Recognition based on Local Binary Patterns with Threshold”, IEEE Conf. in Granular Computing, 2010.
 R. Duda, P. Hart and D. Stock, Pattern Classification, 2nd ed. John Wiley & Sons, 2001.
 H. Jia and A. Martinez, “Support Vector Machines in Face Recognition with Occlusions”, Proc. IEEE Conf. Computer Vision. Pattern Recognition, pp. 136-141, June, 2009.
 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. 236-334, 1965.
 X. He, S. Yan, Y. Hu and P. Niyogi, “Locality Preserving Projections”, Proc. Conf. Advances in neural Information Processing Systems, 2003
 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.
 O. Boiman, E. Shechtman and M. Irani, “In Defense of Nearest-Neighbor Based Image Classification”, IEEE Conf. Computer Vision and Pattern Recognition, 2008.
 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, no. 12, pp. 2037-2041, 2006.
 H. Jia and A. Martinez, “Face Recognition with Occlusions in the Training and Testing Sets”, Proc. IEEE Int. Conf. Autom. Face Gesture Recognition, pp. 1-6, Sep. 2008.
 J. Wright, A. Y. Yang, A. Ganesh, S. S. Sastry and Y. Ma, “Robust Face Recognition via Sparse Representation”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 31, no. 2, pp. 210-227, Feb. 2009.
 A. Martinez. “Recognizing imprecisely localized, partially occluded, and expression variant faces from a single sample per class”, IEEE Trans. PAMI, June 2002.
 X. Tan, S. Chen, Z.-H. Zhou and J. Liu. “Face recognition under occlusions and variant expressions with partial similarity”, IEEE Trans. Information Forensics and Security, June 2009.