成功大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	陳李永 Chen, Lee-Yung
論文名稱：	整合SIFT與PNN於年齡可變之人臉辨識法 Age-Variant Face Recognition Scheme Using Scale Invariant Feature Transform and the Probabilistic Neural Network
指導教授：	李祖聖 Li, Tzuu-Hseng
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系碩士在職專班 Department of Electrical Engineering (on the job class)
論文出版年：	2014
畢業學年度：	102
語文別：	英文
論文頁數：	56
中文關鍵詞：	人臉辨識、機率神經網路、尺度不變特徵轉換
外文關鍵詞：	Face Recognition, PNN, SIFT
相關次數：	點閱：100 下載：3
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

對於人臉老化的差異，如何提高辨識成功率是自動人臉辨識系統中重要的課題。大部分的人臉辨識研究都只專注於老化模擬或年齡估算，但對於有年齡變化的人臉辨識系統而言，需找出一個適合的特徵且設計有效率的匹配框架模型，才能有效提高辨識率。本論文主要在探討因年齡所造成的差異，運用尺度不變特徵轉換(SIFT: Scale Invariant Feature Transform)演算法對於光線、視角改變及雜訊忍受度較高的特性，透過密集採樣進行偵測與描述臉部影像的局部特徵，再利用機率神經網路(PNN: Probabilistic Neural Network)中貝氏決策(Bayes Strategy)來處理分類的問題。此方法藉由調整機率密度函數的平滑參數，即可提高辨識成功率。最後，使用FG-NET (Face and Gesture Recognition Research Network)的人臉資料庫，進行不同年齡的人臉辨識，以驗證此方法有較佳的辨識性能。

Facing to the aging variation problem, how to improve the correct recognition rate of an automatic face recognition system is an important issue. Most face recognition studies only focus on aging simulation or age estimation. For face recognition system under age variation, it is possible to effectively design a suitable and efficient performance matching a framework model. This thesis mainly discusses the differences caused by age level using the Scale Invariant Feature Transform (SIFT) algorithm. Because it has a high tolerance of noise characteristics, the light and viewing angle has changed. It can be detected and can describe local features of the face images through intensively sampling a local descriptor. Then it uses the Probabilistic Neural Network (PNN) by Bayesian classification decisions to deal with the problem by adjusting the smoothing parameter from the probabilistic density function in order to improve the recognition success rate. Finally, the proposed age-variant face recognition scheme is applied to the FG-NET (Face and Gesture Recognition Research Network) face database and the simulation results demonstrate that the correct recognition rate is indeed improved.

Contents

Abstract												   		   	 I
Acknowledgment												    III
Contents														    IV
List of Figures								                     VII
List of Tables                                                X
List of Acronyms												    XI




Chapter 1. Introduction
		1.1 Motivation													1
        1.2 Research												  	3
		1.3 Thesis Organization											9

Chapter 2. The Face Feature Extraction Algorithm
        2.1 Introduction											    10
		2.2 Principal Component Analysis (PCA)						    11
        2.3 Scale Invariant Feature Transform (SIFT)						15
            2.3.1 Scale-space extrema detection			                15
            2.3.2 Accurate key-point localization		                    18
            2.3.3 Orientation assignment                                 20
            2.3.4 Key-point descriptor                                   21
		2.4 Summary													22

Chapter 3. Probabilistic Neural Network (PNN) Classifier 	
3.1 Introduction												24
		3.2 Bayes Classifier                                	      	26
		3.3 Parzen Window Method                                     	27
        3.4 Smoothing Parameter                                        30
        3.5 Design of the Probabilistic Neural Network                      32
        3.6 Architecture of the Probabilistic Neural Network                  33
		3.7 Summary                                                  34

Chapter 4. Experiments
		4.1 Introduction					     			    		36
		4.2 FG-NET Database				                            37
        4.3 Experiments   		  	                        	    39
          4.3.1 Experimental setting	                                    39
          4.3.2 Experimental conditions                                  39
          4.3.3 Output chart examples illustration                          40
          4.3.4 Training test                                            41
		4.4 Experimental Results                                        42
          4.4.1 Setting smoothing parameter                               42
          4.4.2 Results                                                48

Chapter 5. Conclusions and Future Works
		5.1 Conclusions					                			50
		5.2 Future Works			                                    51

References															53

List of Figures
1.1    Face feature match	2
1.2    Face recognition process	3
1.3    Schematic of the aging simulation process from age x to y [20]	4
1.4    Biological neural network by John Wiley & Sons, Inc. 2000.	8

2.1	 2-D images into 1-D vector of diagram	11
2.2	 The input facial image set from ORL database [32]	13
2.3	 (a)The mean-face  (b-e)The eigen-face [32]	14
2.4     Difference of Gaussian (DOG) 	16
2.5     Extrema Detection [22]	17
2.6     Definition of corner and edge region [36] 	20
2.7     Image gradient	21
2.8     Key-point descriptor	21
2.9     Image gradient	22
2.10    Key-point descriptor	22
2.11    Overview of SIFT algorithm	23

3.1     Three different Gaussian functions, and the remaining is the total overlay	28
3.2     Two different two-dimensional Gaussian functions [37]	28
3.3     The architecture of Probabilistic neural network	33
   3.4     Overview of SIFT-PNN Algorithm	35

4.1	FG-NET aging database [38]	37
4.2	From top to bottom (1) angle (2) expression (3) light (4) shelter [38]	38
4.3	Feature extraction by SIFT	40
4.4	Feature match by SIFT	40
4.5	Output data form	41
4.6	PCA-PNN Re-TEST rate 62.65%	41
4.7	SIFT-PNN Re-TEST rate 99.59%	41
4.8	PCA-PNN recognition rate with different smoothing parameters	42
4.9	σ=0.01 of PCA-PNN Scheme	43
4.10	σ=0.03 of PCA-PNN Scheme	43
4.11	σ=0.05 of PCA-PNN Scheme	43
4.12	σ=0.07 of PCA-PNN Scheme	43
4.13	σ=0.09 of PCA-PNN Scheme	43
4.14	σ=0.10 of PCA-PNN Scheme	43
4.15	σ=0.30 of PCA-PNN Scheme	44
4.16	σ=0.50 of PCA-PNN Scheme	44
4.17	σ=0.70 of PCA-PNN Scheme	44
4.18	σ=0.90 of PCA-PNN Scheme	44
4.19	σ=1.00 of PCA-PNN Scheme	44
4.20	σ=5.00 of PCA-PNN Scheme	44
4.21	SIFT-PNN recognition rate with different smoothing parameters	45
4.22	σ=0.01 of SIFT-PNN Scheme	46
4.23	σ=0.03 of SIFT-PNN Scheme	46
4.24	σ=0.05 of SIFT-PNN Scheme	46
4.25	σ=0.07 of SIFT-PNN Scheme	46
4.26	σ=0.09 of SIFT-PNN Scheme	46
4.27	σ=0.10 of SIFT-PNN Scheme	46
4.28	σ=0.30 of SIFT-PNN Scheme	47
4.29	σ=0.50 of SIFT-PNN Scheme	47
4.30	σ=0.70 of SIFT-PNN Scheme	47
4.31	σ=0.90 of SIFT-PNN Scheme	47
4.32	σ=1.00 of SIFT-PNN Scheme	47
4.33	σ=5.00 of SIFT-PNN Scheme	47

List of Tables
1    PCA-PNN recognition rate with different smoothing parameter	42
2    SIFT-PNN recognition rate with different smoothing parameter	45
3    Comparison of age variant face recognition methods on FG-NET database	49
List of Acronyms

BPN	Back Propagation Network
DOG	Difference of Gaussian
EOH	Edge Orientation Histograms
FG-NET	Face and Gesture Recognition Research Network
H	Hessian matrix
HLEE	Hessian Locally Linear Embedding
HOG	Histogram of Oriented Gradient
ICA	Independent Component Analysis
LBP	Local Binary Pattern
LDA	Linear Discriminant Analysis
LE	Laplacian Eigenmaps
LLE	Locally Linear Embedding
MFA	Marginal Fisher Analysis
MFDA	Multi-Feature Discriminative Analysis
PCA	Principal Component Analysis
PDF	Probability Density Function
PNN	Probabilistic Neural Network
SIFT	Scale Invariant Feature Transform



                                    

[1] C. Stauffer and W.E. L. Grimson, ”Learning pattern of activity using real-time tracking, ” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, No. 8, pp. 747-757, Aug. 2000.
[2] C. Y. Lin, P. C. Jo and C. K. Tseng, ”Multi-functional intelligent robot DOC-2,” in Proc, IEEE-RAS Int. Conf. Humanoid Robots, pp. 530-535, Dec. 2006.
[3] U. Ozguner, C. Stiller and K. Redmill, “Systems for safety and autonomous behavior in cars: The DARPA grand challenge experience,” Proc. IEEE, Vol. 95, No. 2, pp. 397-412, Feb. 2007.
[4] Z. Li, U. Park and A. Jain. “A discriminative model for age invariant face recognition.” IEEE Trans. on Information Forensics and Security, Vol. 6, No. 3, pp. 1028-1037, Sep. 2011.
[5] Y. Fu and T. S. Huang, “Human age estimation with regression on discriminative aging manifold,” IEEE Trans. Multimedia, Vol. 10, No.4, pp. 578–584, Jun. 2008.
[6] X. Geng, Z. Zhou and K. Smith-Miles, “Automatic age estimation based on facial aging patterns,” IEEE Trans. Pattern Anal. Machine Intelligence, Vol. 29, No. 12, pp. 2234–2240, Dec. 2007.
[7] G. Guo, Y. Fu, C. Dyer and T. Huang, “Image-based human age estimation by manifold learning and locally adjusted robust regression,” IEEE Trans. Image Process., Vol. 17, No. 7, pp. 1178–1188, Jul. 2008.
[8] G. Guo, G. Mu, Y. Fu and T. Huang, “Human age estimation using bio-inspired features,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 112–119, 2009.
[9] Y. Kwon and N. da Vitoria Lobo, “Age classification from facial images,” Computer Vision Image Understanding, Vol. 74, No. 1, pp. 1–21, 1999.
[10] A. Lanitis, C. Draganova and C. Christodoulou, “Comparing different classifiers for automatic age estimation,” IEEE Trans. System, Man, Cybern., Vol. 34, No. 1, pp. 621–628, Feb. 2004.
[11] A. Montillo and H. Ling, “Age regression from faces using random forests,” in Proc. IEEE Int. Conf. Image Processing, Cairo, Egypt, pp. 2465–2468, 2009.
[12] N. Ramanathan and R. Chellappa, “Face verification across age progression,” IEEE Trans. Image Process., vol. 15, no. 11, pp. 3349–3361,Nov. 2006.
[13] J. Wang, Y. Shang, G. Su and X. Lin, “Age simulation for face recognition,” in Proc. Int. Conf. Pattern Recognition, pp. 913–916, 2006.
[14] S. Yan, H. Wang, X. Tang and T. Huang, “Learning auto-structured regressor from uncertain nonnegative labels,” in Proc. Int. Conf. Computer Vision, pp. 1–8, 2007.
[15] S. Zhou, B. Georgescu, X. Zhou and D. Comaniciu, “Image based regression using boosting method,” in Proc. IEEE Int. Conf. Computer Vision, Vol. 1, pp. 541–548, 2005.
[16] A. Lanitis, C. Taylor and T. Cootes, “Toward automatic simulation of aging effects on face images,” IEEE Trans. Pattern Anal. Machine Intelligence, Vol. 24, No. 4, pp. 442–455, Apr. 2002.
[17] J. Suo, S. Zhu, S. Shan and X. Chen, “A compositional and dynamic model for face aging,” IEEE Trans. Pattern Anal. Machine Intelligence, Vol.32, No. 3, pp. 385–401, Mar. 2010.
[18] J. Suo, X. Chen, S. Shan and W. Gao, “Learning long term face aging patterns from partially dense aging databases,” in Proc. Int. Conf. Computer Vision, pp. 622–629, 2009.
[19] N. Tsumura, N. Ojima, K. Sato, M. Shiraishi, H. Shimizu, H. Nabeshima, S. Akazaki, K. Hori and Y. Miyake, “Image-based skin color and texture analysis/synthesis by extracting hemoglobin and melanin information in the skin,” ACM Trans. Graph., Vol. 22, No. 3, pp. 770–779, 2003.
[20] U. Park, Y. Tong, and A. K. Jain, “Age invariant face recognition,” IEEE Trans. Pattern Anal. Machine Intelligence, Vol. 32, No. 5, pp. 947–954, May. 2010.
[21] 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.
[22] D. G. Lowe, ”Distinctive Image Features From Scale-Invariant Keypoints,” International Journal of Computer Vision, Vol. 60, pp. 91–110, 2004.
[23] N. Dalal and B. Triggs, “Histograms of Oriented Gradients for Human Detection,” IEEE Conference Computer Vision and Pattern Recognition, Vol. 2, pp. 886-893, 2005.
[24] M. Turk, A. Pentland, “Eigenface for recognition,” Journal of Cognitive Neuroscience, Vol. 3, No. 1, pp. 71-86, 1991.
[25] P. Belhumeur, J. Hespanha and D. Kriegman, “Eigenfaces vs. Fisherfaces: Recognition Using Class Spectific Linear Projection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, pp. 711-720, 1997.
[26] P. Comon, “Independent Component Analysis: a new concept?,” IEEE Transactions on Signal Processing, Vol. 36, No 3, pp. 287-314, 1994.
[27] S. Yan, D. Xu, B. Zhang and H. Zhang, “Graph Embedding: A General Framework for Dimensionality Reduction,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 29, No. 1, pp. 40-41, 2007.
[28] J. B. Tenenbaum, V. de Silva and J. C. Langford, ”A Global Geometric Framework for Nonlinear Dimensionality Reduction,” Science, Vol. 290, pp. 2319-2323. 2000.
[29] M. Belkin and P. Nyiogi, ”Laplacian eigenmaps for dimensionality reduction and data Representation,” Neural Computer, Vol. 15, No. 6, pp. 1373-1396, 2003.
[30] S. T. Roweis and L. K. Saul, ”Nonlinear Dimensionality Reduction by Locally Linear Embedding,” Science, Vol. 290, pp. 2323-2326,2000.
[31] D. Donoho and C. Grimes, “Hessian Eigenmaps: Locally Linear Embedding Techniques for High-Dimensional Data,” Proceedings of the National Academy of Sciences, Vol. 100, No. 10, pp. 5591-5596, 2003.
[32] F. Samaria and A. Harter, “Parameterisation of a stochastic model for human face identification,” 2nd IEEE Workshop on Applications of Computer Vision, pp.138-142, 1994.
[33] J. J. Koenderink, “The structure of images,” Biological Cybernetics, Vol. 50, No. 5, pp. 363-396, 1984.
[34] T. Lindberg, ”Scales-space theory: A basic tool for analyzing structures at different scales,” J. Applied Statistics, Vol. 21, No 2, pp. 224-270, 1994.
[35] M. Brown and D. G. Lowe, “Invariant features from interest point groups,” in Proc. British Machine Vision Conf., pp. 656-665, 2002.
[36] C. Harris and M. Stephens, “A combined corner and edge detector,” in Proc. Alvey Vision Conf., pp. 147-151, 1988.
[37] D. F. Specht, “Probabilistic neural networks for classification, mapping or associative memory,＂in Proc. IEEE International Conference on Neural Networks, Vol. 1, pp. 525-532, 1988.
[38] FG-NET Aging Database [Online]. Available: http://www.fgnet resuit.com/
[39] X. Geng, Z. Zhou and K. Smith-Miles, “Automatic age estimation based on facial aging pattern,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 2234-2240, 2007.
[40] C. Harris, “Geometry from visual motion,” Active Vision, pp. 263-284, 1992.
[41] 葉怡成, 類神經網路模式應用與實作, 儒林書局, pp. 2-5, 2003.
[42]　 M. Caudill, “Neural networks primer, Parts I-V,” AI EXPERT, “Part I ” (Dec 1987), pp. 46-52, “Part II ” (Feb 1988), pp. 55-61, “Part III ” (June 1988), pp. 53-59, “Part IV ” (Aug 1988), pp. 61-67, “Part V ” (Nov 1988), pp. 57-64.
[43] R. Hecht Nielsen, “Neurocomputing: picking the human brain,” IEEE Spectrum, pp. 36- 42, March 1988.
[44]　 R. E. Howard, L. D. Jackel, and H. P. Graf, “Electronic neural networks,” AT&T Tech. J., Vol. 67, No. 1, pp.58-64, 1988.
[45]　 C. Y. Tseng, M. S. Chen, “Photo identity tag suggestion using only social network context on large-scale web services,” in Proc. IEEE International Conference on Multimedia and Expo (ICME), pp. 1-4, July. 2011.
[46]　 A. P. James and S. Dimitrijev, “Face Recognition Using Local Binary Decisions,” IEEE Signal Processing Letters, Vol. 15, pp. 821-824, 2008.
[47] A. M. Mood and F. A. Graybill, Introduction to the Theory of Statistics, New York: Macmillan, 1962.
[48] http://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html

校內：2018-09-05公開
校外：2018-09-05公開

簡易檢索 / 詳目顯示

相關論文