跳到主要內容

臺灣博碩士論文加值系統

(3.231.230.177) 您好!臺灣時間:2021/07/28 20:41
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:陳宥竹
研究生(外文):You-Zhu Chen
論文名稱:智慧型機器人頭顱系統設計與實現
論文名稱(外文):Design and Implementation of an Intelligent Robotic Head System
指導教授:蔡清池
學位類別:碩士
校院名稱:國立中興大學
系所名稱:電機工程學系所
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:82
中文關鍵詞:智慧型機器人頭顱
外文關鍵詞:IntelligentRobotic head system
相關次數:
  • 被引用被引用:1
  • 點閱點閱:88
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
本篇論文發展一個擁有臉部偵測、追蹤、辨識以及表情識別的智慧型機器人頭顱的技術。 利用FPGA及伺服馬達來建構一個智慧型的機器人頭顱之後,本論文採用一般標準的影像處理方法來偵測人臉,並利用模糊凱爾曼濾波器(FKF)來進行人臉追蹤 , 再透過主成分分析法(PCA)以及支援向量機(SVM)來進行人臉鑑定和臉部的表情識別。利用電腦模擬以及實驗來加以驗證模糊凱爾曼濾波器(FKF)在人臉追蹤的效果,再透過機器人頭顱系統的實驗來證實臉部辨識及表情識別的效能,這些被提議的技術可能有助於工作服務型機器人領域的專業人士。
This thesis develops techniques for face detection and tracking, face identification and facial expressions recognition of an intelligent robotic head system. After constructing a physical intelligent robotic head using FPGA and RC servomotors, this thesis applies a standard image processing algorithm to detect faces of any users and then track their faces using a fuzzy Kalman filtering scheme. Face identification and facial expressions recognition are achieved by means of principal component analysis (PCA) and support vector machine (SVM). Computer simulations and experimental results are conducted to verify the effectiveness of the proposed face tracking method. The performance and merit of the proposed face identification and face expressions recognition algorithms are exemplified by performing experiments on the experimental robotic head system. The proposed techniques may be of interesting to professionals working in the field of mobile service robots coexisting with people.
Contents
Acknowledgements……………………………………………………...i
Chinese Abstract………………………………………………………..ii
English Abstract………………………………………………………. iii
Contents…………………………………………………………………iv
List of Figures………………………………………………………….vii
List of Tables……………………………………………………………xi
Nomenclature…………………………………………………………..xii

Chapter 1 Introduction..............................................1
1.1 Introduction..........................................1
1.2 Literature Review.....................................3
1.3 Motivation and Objectives.............................5
1.4 Contributions of the Thesis...........................5
1.5 Thesis Organization...................................6
Chapter 2 System Design and Implementation...............7
2.1 Introduction..........................................7
2.2 System Design and Description of AECL-No.1...........10
2.2.1 Hardware design and implementation of the AECL-No.1.....................................................11
2.2.1 Description of small-scare personal computer and touch screen.............................................14
2.2.2 Introduction to Nios II............................17
2.3 FPGA-based PWM generator and the RS-232 serial communication module.....................................20
2.3.1 PWM Signal’s Duty Cycle of RC Motor...............20
2.3.2 RS-232 Transmission Circuit Design.................24
2.3.3 RS-232 Receiver Circuit Design .....................25
2.3.3 MAX232.............................................27
2.4 Distance Calculation Between the Human Face and the Centre of the Binocular Webcams..........................28
2.5 Concluding Remarks...................................30
Chapter 3 Face Tracking Using Fuzzy Kalman Filter.......32
3.1 Introduction.........................................32
3.2 Face Detection .......................................33
3.2.1 YCbCr color model..................................33
3.2.2 Color Skin Separation ..............................34
3.2.2 Edge processing....................................38
3.2.3 Oval model detection...............................39
3.2.3 Extracting features of human faces.................42
3.3 Face Tracking Using Fuzzy Kalman Filter..............43
3.3.1 Kalman Filter......................................43
3.3.2 Exponential Weighted KF............................44
3.3.3 Fuzzy KF (FKF).....................................46
3.3.4 Fuzzy Tuner........................................46
3.3.5 Fuzzy Kalman Filter Algorithm......................52
3.4 Simulation, Experimental Results and Discussion......52
3.4.1 Computer Simulation................................53
3.4.2 Face tracking experiment...........................56
3.5 Concluding Remarks...................................56
Chapter 4 Interactive Face Expression System Design.....59
4.1 Introduction.........................................59
4.2 Harr Wavelet Transform...............................60
4.3 Principal Component Analysis (PCA)...................61
4.4 Recognition Methods..................................63
4.4.1 Support Vector Machines............................63
4.4.1.1 Linear Support Vector Machines...................63
4.4.1.2 Liner Support Vector Machines Non-Separate Case..65
4.4.1.3 Nonlinear Support Vector Machine.................66
4.4.1.4 SVM: Multi-Class Case............................66
4.4.2 Euclidean distance method..........................67
4.5 Interactive Face Expression System Design............68
4.6 Experimental Results and Discussion..................71
4.6.1 Face Identification Experiments and Discussion.....72
4.6.2 Facial Expression Recognition Experiments and Discussion...............................................75
4.7 Concluding Remarks...................................77
Chapter 5 Conclusions and Future Work...................79
5.1 Conclusions..........................................79
5.2 Future Work..........................................80
References...............................................81

List of Figures
Figure 1.1. Picture of EveR-1.............................1
Figure 1.2. Picture of Actroid............................2
Figure 1.3. Picture of Kismet.............................2
Figure 2.1. Pictures of AECL No.1 robotic head and the commercial Robot head WHW8040.............................8
Figure 2.2. System Configuration and signal flow graph of the intelligent head system...............................9
Figure 2.3. System Structure of the AECL-NO.1.............9
Figure 2.4. Pictures of the key components inside the proposed head: (a) Logitech webcam with model of Pro 5000 ; (b) servo-motor...................................12
Figure 2.5. Detailed blue-print of the proposed intelligent head using Solidwork.........................12
Figure 2.6. Pictures of the AECL-No.1....................13
Figure 2.7. Pictures of the Touch screen.................15
Figure 2.8. (a) The front side photograph of XPC.(b) The back side photograph of XPC..............................16
Figure 2.9. Block diagram of the Nios II processor core..16
Figure 2.10. The development flow chart of the Nios II processor system.........................................18
Figure 2.11. Interactive design environment for the use of the IDE..................................................19
Figure 2.12. A real picture of EP1C12....................19
Figure 2.13. Illustration of the angle range of the motor....................................................20
Figure 2.14. Block diagram of the PWM generator..........21
Figure 2.15. Block diagram of the transmitter............21
Figure 2.16. Flow chart of the RS-232 state machine......25
Figure 2.17. Block diagram of the designed RS-232 receiver. ................................................26
Figure 2.18. Software flow chart of the RS-232 receiver. ................................................27
Figure 2.19. Pin Descriptions of IC MAX232...............28
Figure 2.20. The MAX232 converter circuit................28
Figure 2.21. Parameters of a single CCD..................29
Figure 2.22. Binocular stereo vision system..............29
Figure 3.1. Flow chart of the proposed face tracking and detection................................................33
Figure 3.2. The YCbCr color model........................34
Figure 3.3. Fuzzy skin color adjuster....................35
Figure 3.4. Membership functions of the Fuzzy skin color adjuster. (a) Membership function of the input. (b) Membership function of the output........................36
Figure 3.5. Experimental results of the fuzzy skin color adjuster. (a) Normal light intensity. (b) Stronger light intensity. (c) Darker light intensity....................37
Figure 3.6. Experimental pictures of separating skin color. ................................................38
Figure 3.7. (a) A general 3 X 3 mask; (b) mask used to compute vertical response at center point of the 3 X 3 region; (c) mask used to compute horizontal response at center point of the 3 x 3 region.........................39
Figure 3.8.The result by using Sobel edge process........39
Figure 3.9. The Oval template............................40
Figure 3.10. Experimental results of the detection process using the oval model.....................................41
Figure 3.11. Face detection result with oval model using new tactic...............................................42
Figure 3.12. Features extraction of a human face.........43
Figure 3.13. Membership function for the mean of the innovations..............................................49
Figure 3.14. Membership function for the second-order moment of the innovations................................49
Figure 3.15. Membership function for the slew rate of the second-order moment .....................................50
Figure 3.16. The membership function of the weighting factor ..................................................50
Figure 3.17. Block diagram of the fuzzy Kalman filtering algorithm................................................53
Figure 3.18. Comparison of the true position, KF and FKF position estimates.......................................54
Figure3.19. Performance comparison of the KF and FKF estimated states.........................................54
Figure3.20. Comparison of the true position, KF and FKF position estimates.......................................55
Figure 3.21. Comparison of the KF and FKF estimated states. ................................................55
Figure 3.22. Experimental pictures of the face tracking.................................................57
Figure 4.1. Diagram of two-dimensional Harr wavelet transforms...............................................60
Figure 4.2.The results of the Harr wavelet transform; (a) the original image; (b) the result of the Harr wavelet transform; (c) the position relationship picture of sub-image....................................................61
Figure 4.3. A chart of transforming the N by N image to N by 1 vector 62
Figure 4.4. Training Face images. 62
Figure 4.5. Average face . 62
Figure 4.6. The chart of the Support Vector Machines. 63
Figure 4.7. Non-separate case for linear support vector machines. 65
Figure 4.8. The binary tree structure of One-against-one strategy. 67
Figure 4.9. Flow chart of the overall experimental procedure. 68
Figure 4.10. Welcome mode. 69
Figure 4.11. Happy smiling. 70
Figure 4.12. Anger. 70
Figure 4.13. Joy. 70
Figure 4.14. Sadness. 71
Figure 4.15. Database of human faces. 72
Figure 4.16. Harr faces. 72
Figure 4.17. Experimental pictures of face identification. 76
Figure 4.18. Experimental results of facial expression recognition; (a) smiling; (b) anger; (c) sadness; (d) happy smiling. 77

List of Tables
Table 2.1. Comparison of Degrees of Freedoms 8
Table 2.2. Specifications of the AECL No.1. 10
Table 2.3. Specifications of KRS-786ICS and KRS-2350ICS 13
Table2.4. Specifications of the Logitech webcam pro 5000 14
Table 2.5 Specifications of the touch screen 15
Table 2.6. Specifications of the small-scale personal computer 16
Table 2.7. Specifications of the EP1C12. 18
Table 2.8. Relationship between the degree and PWM duty cycle. 22
Table 2.9. Experimental results of the position estimation for the user face. 30
Table 3.1 Fuzzy rules for L 35
Table 3.2. Fuzzy rule matrix for positive slew slope. 51
Table 3.3. Fuzzy rule matrix for zero slew slope. 51
Table 3.4. Fuzzy rule matrix for negative slew slope. 51
Table 3.5 Statistical deviations and mean value of Simulation 1 53
Table 3.6 Statistical dispersion and mean value of Simulation 2 56
Table 4.1 Data of the face identification experiment 73
Table 4.2. Experimental results of the proposed face identification experiment 1 73
Table 4.3 Experimental results of the proposed face identification experiment 2 73
Table 4.4 Data information of the facial expressions recognition experiment 74
Table 4.5.Experimental results of the proposed facial expressions recognition experiment 1……………………………………………………………..74
Table 4.6.Experimental results of the proposed facial expressions recognition experiment 2……………………………………………………………..74

Nomenclature
SOPC System on programmable chip
FPGA Field programmable gate array
DOF Degree of freedom
PWM Pulse width modulation
FKF Fuzzy Kalman filter
Q(K) Covariance matrix of the process noise
R(K) Covariance matrix of the measurement noise
PCA Principal component analysis
SVM Support vector machine
References
[1] A.Albiol, L. Torres, E.J. Delp, “Optimum color spaces for skin detection,” Politechnic University of Valencia Spain, vol.1, , pp.122-124, 2001.
[2] M. Turk and A. Pentland, “Eigenfaces for recognition,” Journal of Cognitive Neuro-science, vol.3, no.1, pp.71-86, 1991.
[3] S. H. Jeng, H. Y. Mark Liao, C. C. Han, M. Y. Chern, and Y. T. Liu, “An efficient approach for facial feature detection using geometrical face model,” in Proc. of the 13th International Conference on Pattern Recognition, Vienna, Austria, 1996, vol.3, pp.426-430.
[4] K. Sobottka and I. Pitas, “Extraction of facial regions and features using color and shape information,” in Proc. 13th International Conference on Pattern Recognition, , Vienna, Austria, Aug. 1996, pp.421-425.
[5] H. Wu, Q. Chen, and M. Yachida, “A fuzzy-theory-based face detector,” in Proc. 13th International Conference on Pattern Recognition, Vienna, Austria, Aug. 1996 pp.406-410.
[6] C. Cortes, V. Vapnik, “Support vector networks,” Machine Learning, vol. 20, pp. 273-297, 1995.
[7] E. Osuna, Support Vector Machines: Training and Applications, Ph.D. thesis, Dept. of EECS, Massachusetts Institute of Technology, 1998.
[8] E. Osuna, R. Freund, and F. Girosi, “Training support vector machines: an application to face detection,” in Proc. Computer Vision and Pattern Recognition, 1997, pp.17-19.
[9] S. Lawrence, C. L. Giles, A. C. Tsoi, and A. D. Back. “Face recognition: A convolutional neural network approach,” in Proc. IEEE Trans. Neural Networks, vol.8, pp.98–113, 1997.
[10] R. Chellappa, C. L.Wilson, and S. Sirohey. “Human and machine recognition of faces: A survey,” in Proc. IEEE, vol.83, pp.705– 741, May 1995.
[11] D. Valentin, H. Abdi., A. J. O’Toole, and G. W. Cottrell, “ Connectionist models of face processing: A survey, ”Pattern Recognition, vol.27 pp.1209–1230, 1994.
[12] A. Samal and P. A. Iyengar, “Automatic recognition and analysis of human faces and facial expressions: A survey,” Pattern Recognition, vol.25, pp.65–77, 1992.
[13]A. G. O. Mutambara, Decentralized estimation and control for multisensor systems, CRC Press LLC, 1998.
[14] L. A. Zadeh, “Fuzzy logic,” Computer, vol. 21, no. 4, pp. 83-93, 1988.
[15] G. N. Saridis, Stochastic processes, estimation, and control: the entropy approach, John Wiley & sons, New York, 1995.
[16]K. Pearson, “On line and planes of closest fit to systems of points in space,” Philosophy Magazine, vol.2, pp.559 - 572, 1901.
[17]H. Hotelling, “of a complex of statical variables into principal components,” Journal of Educational Psychology, vol.24, pp.417 - 441, 1933.
[18] I. T. Jolliffe, Principal Component Analysis, Springer-Verlag, 1986.
[19] V. N. Vapnik. Statistical learning theory. John Wiley & Sons, New York, 1998.
[20] J. Friedman. Another approach to polychotomous classification. Technical report, Department of Statistics, Stanford University, 1996.
[21] U. KreBel, “Pairwise classification and support vector machines,” in proc Advances in Kernel Methods-Support Vector Learning, pp.255-268, Cambridge, MA, 1999, MIT Press.
[22] C.-C. Chang and C.-J. Lin. LIBSVM: a library for support vector machines, 2001. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm.
[24] J. C. Burges, “A tutorial on support vector machines for pattern recognition,” Data Mining and Knowledge Discovery, Vol. 2, pp. 121-167,1998.
[25] S. Haykin, Neural Networks: A Comprehensive Foundation, Second Edition,
New Jersey: Prentice-Hall, 1999.
[26]F. L. Lewis, Optimal estimation with an introduction to stochastic control theory, John Wiley & Sons, 1986.
[27]P. S. Maybeck, Stochastic Models, Estimation, and Control, vol.1, New York, Academic Press, 1979.
[28]J. Z. Sasiadek and Q. Wang, “Sensor fusion based on fuzzy Kalman filtering for autonomous robot vehicle,” in Proceeding the 1999 IEEE Conference on Robotics and Automation, Detroit, Michigan, pp. 2970-2975, May 1999.
[29]J. Z. Sasiadek, and Q. Wang, “Fuzzy adaptive Kalman filtering for INS/GPS data fusion and accurate positioning,” in Proc. of the 2001 IFAC International Symposium on Aerospace Control, Bologna, Italy, pp.451-459,2001.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top