(3.238.96.184) 您好!臺灣時間:2021/05/12 22:36
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:黃雅紫
研究生(外文):Ya-Tzu Huang
論文名稱:混合田口式粒子群最佳化演算法為基礎之類神經網路於二維/三維人臉辨識應用
論文名稱(外文):2D/3D Face Recognition Using Neural Networks Based on Hybrid Taguchi-Particle Swarm Optimization
指導教授:林正堅林正堅引用關係
指導教授(外文):Cheng-Jian Lin
學位類別:碩士
校院名稱:朝陽科技大學
系所名稱:資訊工程系碩士班
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2008
畢業學年度:96
語文別:英文
論文頁數:55
中文關鍵詞:粒子群最佳化主成份分析多層類神經網路人臉辨識田口方法賈伯小波。
外文關鍵詞:multilayer neural networks (MLNN)Gabor waveletFace recognitionparticle swarm optimizationprincipal component analysis (PCA)Taguchi method
相關次數:
  • 被引用被引用:0
  • 點閱點閱:357
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:4
  • 收藏至我的研究室書目清單書目收藏:0
近年來有許多基於電腦視覺的系統變得越來越重要,其應用如監控系統、自動存取控制系統與人機互動等等。然而,在應用中,人臉辨識系統將扮演著重要的角度,不過對於影像受光源變化、人臉姿態與表情改變時會使得人臉辨識變得越來越困難。因此,在本篇論文中我們利用臉部紋理和表面資訊提出一人臉辨識的方法。所提出的方法不但可以用來處理臉部姿態與表情改變的問題,並且改善人臉辨識效能。我們首先以灰階的臉部影像使用賈伯小波濾波器在不同大小和定向遮罩中擷取局部特徵,然後基於主成分分析(PCA)從灰階影像和面部的表面影像獲得特徵向量,結合紋理與表面特徵向量。至於在辨識部分,我們提出一個基於多層類神經網路的混合式田口粒子群演算法(HTPSO)以完成人臉辨識。實驗結果由十個人不同的臉部姿態與表情以證明我們方法的效能。另外,將我們所提出學習方法與其他常見的方法相比較,例如倒傳遞演算法(BP),粒子群最優化演算法(PSO)以及遺傳基因演算法(GA),並且評估不同的資料形態。最後,實驗結果證明所提出的HTPSO學習演算法在辨識率方面可以有效優於其它方法。
Mary computer vision-based systems have become more and more important in recent years, such as the surveillance, automatic access control and the human-robot interaction. Face recognition plays a critical role in those applications. However, face recognition is a very difficult problem due to a substantial variation in light direction, different face poses, and diversified facial expressions. Therefore, we present a method of face recognition using facial texture and surface information. This method deals with face pose and facial expression problems of changes, and improves the recognition performances. We first use Gabor wavelets extracting local features at different scales and orientations by gray facial images, then combine the texture with the surface feature vectors based on principal component analysis (PCA) to obtain feature vectors from gray and facial surface images. We propose a hybrid Taguchi particle swarm optimization (HTPSO) algorithm for face recognition based on multilayer neural networks as an identification model. Experimental results demonstrate the efficiency of our method for 10 individuals with different face poses and facial expressions. In addition, our work compared with other proposed approaches such as back-propagation (BP), particle swarm optimization (PSO) and the genetic algorithm (GA). With different data modality the experimental results demonstrated that the proposed HTPSO learning algorithm is better than the other approaches in recognition rates.
Contents
摘要....................................................................................................................... I
Abstract ...............................................................................................................III
Acknowledgment ............................................................................................. V
Contents...............................................................................................................VI
List of Figures ................................................................................................ VIII
List of Tables.................................................................................................... X
Chapter 1 : Introduction..................................................................................... 1
1.1 Motivation ............................................................................................... 1
1.2 Literature Review................................................................................... 2
1.3 Thesis Organization................................................................................ 5
Chapter 2 : Previous Work ................................................................................ 7
2.1 Facial Images Acquisition...................................................................... 7
2.2 Locating the Feature Points .................................................................. 8
2.3 Data Normalization ................................................................................ 9
2.4 Converting to Gray Images ................................................................. 10
Chapter 3 : Image Analysis .............................................................................. 11
3.1 Gabor Wavelet ...................................................................................... 11
3.2 Wavelet Transform .............................................................................. 15
3.3 Principal Component Analysis............................................................ 16
Chapter 4: Neural Network Classifier ............................................................ 20
Chapter 5 : The Proposed Method .................................................................. 24
5.1 Review of Taguchi Method.................................................................. 24
5.2 A Hybrid Taguchi Particle Swarm Optimization Algorithm .......... 29
Chapter 6: Experimental Results .................................................................... 37
Chapter 7 : Conclusions.................................................................................... 48
Bibliography....................................................................................................... 50
Vita...................................................................................................................... 55

List of Figures
Figure 1-1. Schematic structures of face recognition. ·····································5
Figure 2-1. The digital 3D image camera and some exampled original
facial image. (a) Input device (b) Example of original facial. ·························7
Figure 2-2. The landmark point L1-L10 on face. ············································8
Figure 2-3. Extracting a new facial surface and color image, and the image
is a regular grid of 100*100 pixels.·································································9
Figure 2-4. The subject’s face under different facial expressions and also
the different subjects under the same expression. (a) Range images of a
subject’s face under different facial expressions. (b) Images of three
different subjects under the same expression. ···············································10
Figure 3-1. Combine both the real part and the imaginary parts. (a) Gaobr
Wavelets of Gabor kernels at five scales and eight orientations. (b) Real
part of the convolution outputs in face image. (c) The feature vector
consists of the real and the imaginary part of the Gabor transform.··············14
Figure 3-2. Schematic diagram of wavelet decomposition.···························15
Figure 3-3. A matrix division of the high and low frequency results. ···········16
Figure 4-1. Three-layer feed-forward neural network. ··································21
Figure 5-1. Response graphs. ········································································29
Figure 5-2. Schematic diagram of the swarm.···············································30
Figure 5-3. Flowchart of the proposed HTPSO learning process.·················32
Figure 5-4. Recognition rate of different Vmax value.·····································34
Figure 6-1. The face database includes gray and depth image using facial
images of 10 individuals in Chaoyang University of technology, the data
modality is divided as: (a)Gray images. (b) Depth images.···························40
Figure 6-2. The gray and surface image of a person used for training and
testing respectively.·······················································································41
Figure 6-3. Performance of recognition using the different methods. (a)
Data type is gray information (b) Data type is depth information.················45
Figure 6-4. Compare performance of the different data type. ·······················46
Figure 6-5. Best strategy parameters of three sample modality. (a)Use gray
information. (b) Use depth information. (c)Use 3D information (combine
the gray and depth information). ···································································47

List of Tables
Table 2-1. Numbers and names of landmark point used in 3D face.···············8
Table 5-1. Levels of different parameters. ····················································28
Table 5-2. Matrix experiments with L4(23) OA.············································28
Table 5-3. SNR of each operating parameter and level.································28
Table 5-4. Recognition rate of Different Vmax value. ·····································33
Table 5-5. Factors and the corresponding parameters. ··································34
Table 5-6. Levels of different parameters. ····················································34
Table 5-7. Matrix experiments with L16 (45) OA.··········································35
Table 6-1. Performance of recognition using various methods in gray
database.········································································································43
Table 6-2. Performance of recognition using various methods in depth
database.········································································································43
Table 6-3. Comparison results of the different type image data using
HTPSO method. ····························································································46
[1]C. Liu, “Gabor-based kernel PCA with fractional power polynomial models for face recognition,” IEEE Transactions on pattern analysis and machine intelligence, 2004, Vol. 26, Issue 5, pp.572-581.
[2]M. Oravec, and J. Pavlovicova, “Face recognition methods based on principal component analysis and feedforward neural networks,” IEEE International Joint Conference on Neural Networks, 2004, Vol. 1, pp. 25-29.
[3]J. Kim, J. Choi and M. Turk, “Effective representation using ICA for face recognition robust to local distortion and partial occlusion,” IEEE transactions on pattern analysis and machine intelligence, 2005, Vol. 27, Issue 12, pp.1977-1981.
[4]D. Zhou, X. Yang, N. Peng and Y. Wang, “Improved-LDA based face recognition using both facial global and local information,” Pattern Recognition Letters, 2006, Vol.27, Issue 6, pp 536–543.
[5]Y. He, L. Zhaoa, and C. Zoua, “Face recognition using common faces method,” Pattern recognition, 2006, Vol. 39, Issue 11, pp. 2218-2222.
[6]P. C. Yuen, D. Q. Dai, and G. C. Feng, “Wavelet-based PCA for human face recognition,” IEEE Southwest Symposium on Image Analysis and Interpretation, 1998, Vol. 5-7, pp. 223-228.
[7]P. C. Yuen , and D. Q. Dai, and G. C. Feng, “Face recognition by applying wavelet subband representation and kernel associative memory,” IEEE Transactions on neural networks, 2004, Vol. 15, pp 166-177.
[8]C. J. Lin, and C. C. Chin, “Prediction and identification using Wavelet-based recognition fuzzy neural networks.” IEEE Transactions on Systems, Man, and Cybernetics (Part:B), 2004, Vol. 34, No. 5, pp. 2144-2154.
[9]G. Pan, S. Han, Z. Wu and Y. Wang, “3D Face Recognition using Mapped Depth Images,” 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005, Vol. 3, No. 20-26, pp. 175 – 175.
[10]X. Lu and A. K. Jain, “Automatic feature extraction for multiview 3D face recognition,” Proceedings of the 7th International Conference on Automatic Face and Gesture Recognition, 2006, pp. 585-590.
[11]S. Chafik, S. Anuj, and D. Mohamed, “Three-Dimensional face recognition using shapes of facial curves,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, Vol. 28, No. 11, pp. 1858-1863.
[12]C. J. Lin and Y. J. Xu, ” A self-constructing neural fuzzy network with dynamic-form symbiotic evolution,” International Journal of Intelligent Automation and Soft Computing, 2007, Vol. 13, No. 2, pp. 123-137.
[13]C. J. Lin and Y. J. Xu, ”A novel genetic reinforcement learning for nonlinear fuzzy control problems,” Neurocomputing, 2006, Vol. 69, Issues. 16-18, pp. 2078-2089.
[14]M. S. Obaidat and D. T. Macchiarolo, “A multilayer neural network system for computer access security,” IEEE Transactions on Systems, Man, and Cybernetics, 1994, Vol. 24, No. 5, pp. 806-813.
[15]R.C. Lacher, S. I. Hruska, and D. C. Kuncickly, “Back-propagation learning in expert networks,” IEEE Transactions on Neural Networks, 1992, Vol. 3, No.1, pp.62-72.
[16]C. F. Juang, “A hybrid of genetic algorithm and particle swarm optimization for recurrent network design,” IEEE Transactions on System, Man and Cybern, 2004, Vol. 34, Issue 2, pp.997–1006.
[17]B. Liu, L. Wang and Y.-H. Jin, “An effective PSO-based memetic algorithm for flow shop scheduling,” Man and Cybernetics, Part B, IEEE Transactions on Systems, 2007, Vol. 37, Issue 1, pp.18–27.
[18]M. I. Chacon and P. Rivas-Perea, “Performance analysis of the feedforward and SOM neural networks in the face recognition problem,” IEEE Symposium on Computational Intelligence in Image and Signal Processing, 2007, pp.313–318.
[19]R. Mendes, P. Cortez, M. Rocha, and J. Neves, “Particle swarms for feedforward neural network training,” The 2002 International Joint Conference on Neural Networks, 2002, pp. 1895-1899.
[20]J. Kennedy, and R. Eberhart, “ Particle swarm optimization,” IEEE Conference on Neural Networks, 1995, Vol. 4, pp. 1942-1948.
[21]C. J. Lin and Y. J. Xu, ”The design of TSK-type fuzzy controllers using a new hybrid learning approach,” International Journal of Adaptive Control and Signal Processing, 2006, Vol. 20, Issue 1, pp. 1-25.
[22]Y. S. Goh and E. C. Tan, “A multilayer neural network system for computer access security,” IEEE Region 10''s Ninth Annual International Conference, 1994, Vol.2, pp.801–804.
[23]Y. Shi, R.C. Eberhart, “Empirical study of particle swarm optimization,” Proceedings of the 1999 Congress on Evolutionary Computation, 1999, pp.1945-1950.
[24]T.P. Bagchi, Taguchi Methods Explained—Practical Steps to Robust Design. Prentice Hall of India, 1993.
[25]P.J. Ross, Taguchi Techniques for Quality Engineering. McGraw Hill, 1996.
電子全文 電子全文(本篇電子全文限研究生所屬學校校內系統及IP範圍內開放)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊
 
系統版面圖檔 系統版面圖檔