Recognition of Painful Facial Expression using Multiple Kernel Learning

簡易檢索 / 詳目顯示

回結果列表

研究生：	王佳琪 Wang, Chia-Chi
論文名稱：	Recognition of Painful Facial Expression using Multiple Kernel Learning 基於多重核心學習方法之痛苦表情辨識
指導教授：	陳永昌 Chen, Yung-Chang
口試委員:	賴文能林惠勇
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2011
畢業學年度：	99
語文別：	英文
論文頁數：	48
中文關鍵詞：	人臉定位、多重核心學習方法
外文關鍵詞：	face detection, Multiple Kernel Learning
相關次數：	點閱：1 下載：0
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

近年來隨著人口老化及少子化的因素，人口組成逐漸邁入高齡化，如何提供老年人看護系統，已是一個重大的課題。我們希望可以利用最近幾年發展的人臉偵測以及人臉辨識的技術發展出以辨識疼痛表情為主的技術，希望藉由對於人臉痛苦表情的分析，能夠在使用者無法明確下達指令時，還能辨認出需要幫助的人並給予協助。
本篇論文利用人臉定位技術Active Appearance Model方法做人臉特徵的定位，然後根據人臉部特徵的特性，將Active Appearance Model定位的特徵點做調整，確保特徵點不會受頭部旋轉動作的影響，跟臉部特徵不會受光影變化影響，之後利用觀察痛苦表情變化的特性做特徵的擷取，我們擷取的特徵除了形狀的特徵外，還包含外觀上的特徵，因為在表現痛苦的表情中，我們可以發現因扭曲表情而產生的皺紋是很重要的特徵。先以人工分類一些有疼痛表情的圖像與沒有疼痛表情的圖像，之後利用改進Support Vector Machine方法的Multiple Kernels Learning方法當分類的系統，用先前已經分類的圖像為基準去產生分類的模型。
最後在實驗結果的部分，我們做了單張圖片的痛苦表情分類，以及影像的痛苦表情分類。由實驗的結果可以看到我們的分類模型，可以辨識出因疼痛而大叫的臉部表情，以及忍耐疼痛的表情，在特徵表現上可以看到這兩種表情是不太相同的，而利用我們的分類模型可以辨識出來。

Nowadays, along with the aging of population and declining birthrate factors, how to provide the elderly care system is a major issue. We want to use the technology of face detection and facial expression recognition which have been developed in recent years to identify facial expressions with pain. If we could detect the pain expression, it may help people who are in accident, such as heart attack.
This thesis uses Active Appearance Model method to detect location of face features. Then by using the feature points of Active Appearance Model, we extract feature points as shape feature then eliminate effect of head rotation in shape feature. We use the feature points of Active Appearance Model to select region of face, then use variance of regions to determine texture feature. Final, we use the Multiple Kernels Learning method as a classifier, which improves the problem about choosing a best kernel in Support Vector Machine. By using Multiple Kernels Learning method, we can easily control the features and know the importance of each kind of features.
Finally, in our experimental results, we show not only recognizing facial expressions with pain in a single image, but also recognizing facial expressions with pain in a video. We can identify the facial expression that a patient feels pain and scream, and the facial expression that a patient feels pain and bear it. That is totally different, but both types of pain are recognized.

Chapter 1: Introduction    1
        1.1 Motivation    1
        1.2 Thesis Organization 2

Chapter 2: Related Works and System Overview    3
        2.1 Overview    3
        2.2 Related Works    3
           2.2.1 Face Feature Extraction        4
           2.2.2 Recognition of Face Expression 7
        2.3 System Overview    10

Chapter 3: Feature Extraction    12
        3.1 Overview    12
        3.2 Active Appearance Model    12
        3.3 Shape Feature    14
        3.4 Texture Feature    18



Chapter 4: Classification System - MKL    22
        4.1 Multiple Kernel Learning    22
        4.2 Theory of Multiple Kernel Learning    23
        4.3 Optimal Multiple Kernel Learning    27
        4.3 Summary     32

Chapter 5: Experimental Results    33
        5.1 Data sets    34
        5.2 Experimental steps    35
        5.3 Recognition of painful face expression  39

Chapter 6: Conclusion and Future Works    44
        6.1 Conclusion    44
        6.2 Future Works      45

References    47








List of Figures

Figure 2.1: Action Units    8
Figure 2.2: Representing FAPs    9
Figure 2.3: The Flow Chat of Training Part    11
Figure 2.4: The Flow Chat of Testing part    11
Figure 3.1: Head Rotation    15
Figure 3.2: Face Rotation    16
Figure 3.3: The representation of symmetric feature pair.    17
Figure 3.4: Gamma correction    19
Figure 3.5: The changed regions when face turns left    21
Figure 4.1: Support Vector Machine    24
Figure 4.2: Flow Chart of Multiple Kernel Learning    25
Figure 4.3: Representing Different Features    26
Figure 5.1: The Illustration of Eye Feature Extraction    36
Figure 5.2: The Illustration of Mouth Feature Extraction    37
Figure 5.3: the Flow Chat of Training Steps    39
Figure 5.4: The Result 1 of Sequence-level     41
Figure 5.5: The Result 2 of Sequence-level    42






List of Tables

Table 5.1: Experimental steps    41
Table 5.2: The Ratio of Error and Accuracy    49
Table 5.3: Type of Classifies    51

                                

[1]Ming-Hsuan Yang, David Kriegman, and Narendra Ahuja, "Detecting Faces in Images: A Survey", IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), vol. 24, no. 1, pp. 34-58, 2002

[2]G. Yang and T. S. Huang, “Human Face Detection in Complex Background,” Pattern Recognition, vol. 27, no. 1, pp. 53-63, 1994.

[3]C. Kotropoulos and I. Pitas, “Rule-Based Face Detection in Frontal Views,” Proc. Int’l Conf. Acoustics, Speech and Signal Processing, vol. 4, pp. 2537-2540, 1997.

[4] S.A. Sirohey, “Human Face Segmentation and Identification,”Technical Report CS-TR-3176, Univ. of Maryland, 1993.

[5] J. Canny, “A Computational Approach to Edge Detection,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 8, no. 6, pp. 679-698, June 1986.

[6] R.L.Hsu, “Face Detection and Modeling for Recognition,” Ph.D dissertation, Michigan State University, East Lansing, MI, United States of America, 2002.

[7] J. Miao, B. Yin, K. Wang, L. Shen, and X. Chen, “A Hierarchical Multiscale and Multiangle System for Human Face Detection in a Complex Background Using Gravity-Center Template,” Pattern Recognition, vol. 32, no. 7, pp. 1237-1248, 1999.

[8] A. Lanitis, C.J. Taylor, and T.F. Cootes, “An Automatic Face Identification System Using Flexible Appearance Models,” Image and Vision Computing, vol. 13, no. 5, pp. 393-401, 1995.

[9] M. Turk and A. Pentland, “Eigenfaces for Recognition,” J. Cognitive Neuroscience, vol. 3, no. 1, pp. 71-86, 1991.

[10] Viola and Jones, "Rapid object detection using boosted cascade of simple features", Computer Vision and Pattern Recognition, 2001

[11]COOTES, T. F., EDWARDS, G. J., AND TAYLOR, C. J. 2001. Active appearance models. IEEE Trans. Patt. Anal. Mach. Intell. 23, 681–685.

[12]M. Kass, A. Witkin, and D. Terzopoulos, “Snakes: Active Contour Models,” Proc. First IEEE Int’l Conf. Computer Vision, pp. 259-269, 1987.

[13]P. Ekman and W.V. Friesen, “Manual for the Facial Action Coding System,” Consulting Psychologists Press, 1977.

[14] F. Bourel, C.C. Chibelushi and A. A. Low, “Recognition of Facial Expressions in the Presence of Occlusion,” Proc. of the Twelfth British Machine Vision Conference, vol. 1, pp. 213–222, 2001.

[15] M. Pardas and A. Bonafonte, “Facial animation parameters extraction and expression recognition using Hidden Markov Models,” Signal Processing: Image Communication, vol. 17, pp. 675–688, 2002.

[16] P.S. Aleksic, A.K. Katsaggelos, “Automatic Facial Expression Recognition Using Facial Animation Parameters and MultiStream HMMs,” IEEE Trans. Information Forensics and Security, vol. 1, no. 1, pp. 3-11, 2006.

[17] P. Michel and R. Kaliouby, “Real Time Facial Expression Recognition in Video Using Support Vector Machines,” Proc. 5th Int. Conf. Multimodal Interfaces, Vancouver, BC, Canada, pp. 258–264, 2003.

[18]Chih-Chung Chang and Chih-Jen Lin, LIBSVM : a library for support vector machines, 2001. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm

[19]Wächter and L. T. Biegler, On the Implementation of a Primal-Dual Interior Point Filter Line Search Algorithm for Large-Scale Nonlinear Programming, Mathematical Programming 106(1), pp. 25-57, 2006

[20]Rakotomamonjy, A., Bach, F., Canu, S., & Grandvalet, Y. (2008). SimpleMKL. J. Mach. Learn. Res., 9, 2491--2521.

[21]P. Lucey, J.F. Cohn, K.M. Prkachin, P.E. Solomon and I. Matthews, "PAINFUL DATA: The UNBC-McMaster Shoulder Pain Expression Archive Database", submitted to the IEEE International Conference on Automatic Face and Gesture Recognition (FG2011), Santa Barbara, USA, 2011.

全文公開日期本全文未授權公開 (校內網路)
全文公開日期本全文未授權公開 (校外網路)

簡易檢索 / 詳目顯示

相關論文