有鑒於近年人口高齡化以及低出生率的現象越趨明顯，獨居問題 也慢慢浮上檯面，如何能夠讓上了年紀的長者或是獨居的人們也能夠 保持身體的健康，類似的課題一再地被提出討論。再加上現代人由於生 活過於緊張忙碌，常常因為長時間維持不正當的姿勢導致身體非常容 易出現狀況，其實被這些問題困擾的人們只要適當的運動就能夠排除 大部分的不適感。並且因為這幾年 COVID-19 疫情的發展，許多原本 會去醫療院所進行復健的患者只能自行在家進行類似的復健運動，但 是他們還是需要正確的引導才能獲得有效的成果。對於上述的對象該 如何提供正確的引導協助，這也是當前許多研究者所關注的議題。

本論文主要針對需要在居家進行運動的對象設計一套輔助系統， 讓這些對象在家裡也能維持基本動作的練習。系統中透過擴增實境技 術將虛擬教練結合正確的指引路徑顯示在螢幕上，使用者可先透過觀 看教練的動作後再跟隨路徑完成正確的動作。為了使本研究中提出的 系統可以提供更穩定的服務品質，在網路環境上將透過雙路由防火牆 備援機制維持伺服器端的正常穩定網路連線，並且透過使用者之間以 混合式 P2P 檔案分享的方式取得較大的影像檔，藉此降低伺服器的網 路連線需求;在影像處理部分則除了提供動作引導的影像路徑外，亦會 藉由動作捕捉進行姿態辨識判斷使用者的操作是否正確。希望能夠透 過本研究提出的架構，讓使用者在家以能進行一些基本的運動練習。

In view of the aging population and declining birthrate in recent years, the problem of living alone has gradually surfaced. How to keep the elderly or people living alone in good health is a popular topic. Because modern people's life is too busy, it is often caused by maintaining improper posture for a long time that the body is very prone to problems, such as back pain, shoulder, and neck pain, etc. In fact, people suffering from these problems can get rid of most of the discomfort with just proper exercise. The development of the COVID-19 epidemic in recent years has also brought great changes. Many patients who would otherwise go to a medical facility for rehabilitation can only perform similar rehabilitation exercises at home, but they still need the right guidance to achieve effective results. How to provide correct guidance and assistance for the above-mentioned objects is also a topic that many researchers are currently concerned about.

This thesis mainly designed an assistance system for users who need to exercise at home, and these users can maintain the training of basic movements at home to improve their physical condition. The system will give correct motion guidance through the virtual coach image in augmented reality. To enable the system proposed in this study to provide a more stable quality of service, the dual-route firewall backup mechanism will be used to maintain the normal and stable network connection of the server in the network environment, and the hybrid P2P file sharing will be used between users. In the image processing part, in addition to providing a motion-guided image path, it also judges the user's operation by his pose estimation is it right or not. It is hoped that the framework proposed in this study will allow users to perform some basic exercise exercises at home.

Abstract i
摘要 iii
致謝 v
Contents vii
Figures ix
Tables xi
1. Introduction 1
1.1 Motivation 1
1.2 Goals 2
1.3 Contribution 3
1.4 Thesis outline 4
2. Background knowledge 5
2.1 Depth sensor camera 5
2.2 Douglas-Peucker algorithm 6
2.3 Integral image 7
2.4 Color space 8
3. Related works 11
3.1 Body recognition & tracking 11
3.2 Face detection 13
4. Research Methods 17
4.1 Design of image processing 18
4.1.1 Body recognition & tracking 20
4.1.2 Face tracking 30
4.2 Design of network 43
4.2.1 Dual-Route backup firewall service 44
4.2.2 HybridP2P 52
5. Experiment result 59
5.1 Dual-Route backup firewall 59
5.2 Body recognition & tracking 62
5.3 Face tracking 66
6. Simulation 71
6.1 Simulation 1 71
6.2 Simulation 2 73
7. Conclusion and future works 79
Reference 83

[1] Jamie Shotton, Andrew Fitzgibbon, Mat Cook, Toby Sharp, Mark Finocchio, Richard Moore, Alex Kipman, Andrew Blake, “Real-time human pose recognition in parts from single depth images.” In Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR '11). IEEE Computer Society, Washington, DC, USA, 1297-1304, 2011.
[2] Dimitrios Alexiadis, Philip Kelly, Tamy Boubekeur, Maher Ben Moussa, “Evaluating a dancer’s performance using kinect-based skeleton tracking.” In: Proceedings of the 19th ACM International Conference on Multimedia, pp. 659–662, 2011.
[3] Paisarn Muneesawang, Naimul Mefraz Khan, Mattew Kyan, R. Bruce Elder, Nan Dong, Guoyu Sun, Haiyan Li, Ling Zhong, Ling Guan, “A machine intelligence approach to virtual ballet training.” IEEE Multimedia 2015; 22(4): 80–92, 2015.
[4] Oikonomidis I., N. Kyriazis and A. Argyros, “Efficient model-based 3-D tracking of hand articulations using Kinect.” Proc. of the 22nd British Machine Vision Conference, BMVC’2011, UK, pp.101.1– 101.11, 2011.
[5] Dutta T., “Evaluation of the KinectTM sensor for 3-D kinematic measurement in the workplace.” Appl Ergon. July 2012.
[6] David H. Douglas and T. Peucker, “Algorithms for the reduction of the number of points required to represent a line or its caricature.” Canadian Cartographer, 10(2):112-122, 1973
[7] D. Bradley, G. Roth, “Adaptive Thresholding Using the Integral Image.” J. Graphics Tools, vol. 12, no. 2, pp. 13-21, 2007.
[8] K. Gunadi, Liliana and J. Tjitrokusmo, “Fall detection application using Kinect”, Proc. Int. Conf. Soft Comput. Intell. Syst. Inf. Technol., pp. 279-282, 2017.
[9] Hongjian Liao and Zhe Qu, “Virtual experiment system for electrician training based on Kinect and Unity3D”, Mechatronic Sciences Electric Engineering and Computer (MEC) Proceedings 2013 International Conference on, pp. 2659-2662, 2013.
[10] M. Kyan, G. Sun, H. Li, L. Zhong, P. Muneesawang, N. Dong, et al., “An approach to ballet dance training through MS Kinect and visualization in a CAVE virtual reality environment”, ACM Trans. Intell. Syst. Technol., vol. 6, no. 2, pp. 1-37, May 2015.
[11] Pedraza-Hueso, M., Martín-Calzón, S., Díaz-Pernas, F. J., and Martínez-Zarzuela, M., “Rehabilitation using kinect-based games and virtual reality”, Procedia Computer Science, 75, 161-168, 2015.
[12] N. Li, S. Zhou, Z. Xiao and Y. Zhao, “Key Techniques of Upper Limb Rehabilitation Training System based on Kinect”, 2020 International Conference on Virtual Reality and Visualization (ICVRV), pp. 171-178, 2020.
[13] A. S. Faradyza, S. Sendari, A. N. Handayani, M. Efendi, J. S. Saputra, I. A. E. Zaeni and O. Fukuda, “Real Time Gesture Detection Using Kinect in Rehabilitation Therapy for Children with Disability”, 2021 7th International Conference on Electrical, Electronics and Information Engineering (ICEEIE), pp. 452-456, 2021.
[14] Nikolaidis and I. Pitas, “Facial feature extraction and pose determination”, Pattern Recognition, Elsevier, vol. 33, no. 11, pp.1783-1791, 2000.
[15] I. Aldasouqi and M. Hassan, “Smart human face detection system”, International Journal of Computers, vol. 5, no. 2, pp. 210-216, 2011.
[16] G. Kim, J. Suhr, H. Jung and J. Kim, “Face occlusion detection by using B-spline active contour and skin color information”, In proceeding of the 11th International Conference. Control, Automation, Robotics and Vision (ICARCV), Singapore, pp. 627-632, 2010.
[17] M. Heshmat, M. Girgis, W. M. Abd-Elhafiez and S. Elaw, “An efficient scheme for face detection based on contours and feature skin recognition”, 2015 Tenth International Conference on Computer Engineering & Systems (ICCES), pp. 255-260, 2015.
[18] Zhang Ning and Geng Xutao, “Face detection based on skin color extraction scheme”, IOP Conf. Ser.: Mater. Sci. Eng., 2019.
[19] Y. -C. Yi and G. Feng, “Face Detection Algorithm with Multi-stage Decision Based on Skin-color and Geometrical Features,” 2018 12th IEEE International Conference on Anti-counterfeiting, Security, and Identification (ASID), pp. 39-42, 2018.
[20] Wilson P.I. and Fernandez J., “Facial feature detection using Haar classifiers.” J. Comput. Sci. Coll. 21, 127-133, 2006.
[21] Takeshi Mita, Toshimitsu Kaneko, and Osamu Hori, “Joint Haar-like Features for Face Detection.” In Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2 (ICCV '05), Vol. 2. IEEE Computer Society, Washington, DC, USA, 1619-1626, 2005.
[22] Hsin-Hung Chiang, Wei-Ming Chen, Han-Chieh Chao, De-Li Tsai. “A virtual tutor movement learning system in eLearning”, Multimedia Tools and Applications, Vol. 78, Issue 4, pp. 4835-4850, February 2019.
[23] Mostafa Karbasi, Zeeshan Bhatti, Parham Nooralishahi, Asadullah Shah, Seyed Mohammad Reza Mazloomnezhad. “Real-Time Hands Detection in Depth Image by Using Distance with Kinect Camera. International Journal of Internet of Things, 4(1A), 1–6, 2015.
[24] Hsin-Hung Chiang, Wei-Ming Chen, Chiou-Shan Chou, Han-Chieh Chao, “Real-time patients' face tracking based on facial feature matching,” 2012 International Symposium on Intelligent Signal Processing and Communications Systems, pp. 569-573, 2012.
[25] Hsin-Hung Chiang, Wei-Ming Chen, Chiou-Shan Chou, Han-Chieh Chao, “Real-time Patients’ Face Tracking Based on Facial Feature Matching”, Journal of Applied Science and Engineering, Vol. 16, No. 3, September 2013.
[26] Xavier P. Burgos-Artizzu and Pietro Perona. Robust face landmark estimation under occlusion. In International Conference on Computer Vision, pages 1513–1520, 2013.
[27] P. Viola and M. Jones, “Rapid Object Detection using a Boosted Cascade of Simple,” Proc. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol. 1, pp. 511, 2001.
[28] T. Minagawa, H. Saito, and S. Ozawa, “Face-Direction Estimating System Using Stereo Vision,” Proc. International Conference on Industrial Electronics, Control and Instrumentation, pp. 1454-1459, 1997.
[29] B. A. Maxwell, R. M. Friedhoff, and C. A. Smith, “A bi-illuminant dichromatic reflection model for understanding images”, Proc. IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-8, 2008.
[30] Soriano, M., B. Martinkauppi, S. Huovinen and M. Laaksonen, “Skin Color Modeling under Varying Illumination Conditions Using the Skin Locus for Selecting Training Pixels,” Proc. IEEE Nordic Signal Processing Symposium, Kolmården, Sweden, 2000.
[31] M. Soriano, B. Martinkauppi, S. Huovinen and M. Laaksonen, “Skin Detection in Video under Changing Illumination Conditions,” Proc. 15th International Conference on Pattern Recognition, Barcelona, Spain, 2000.
[32] R. C. Gonzalez and R. E. Woods, Digital Image Processing, 2nd ed. Prentice Hall Press, 2002.
[33] C. Garcia and G. Tziritas, “Face Detection Using Quantized Skin Color Regions Merging and Wavelet Packet Analysis”, Proc. IEEE Trans. On Multimedia, Vol. 1, No. 3, pp. 264-277, 1999.
[34] A. P. Chen, L. A. Pan, Y. B. Tong, and N. Ning, “Face Detection Technology Based on Skin Color”, Proc. Second International Workshop on Education Technology and Computer Science, Vol. 2, pp. 708-711, 2010.
[35] K. Kryszczuk and Andrzej Drygaj, “Color Correction for Face Detection Based on Human Visual Perception Metaphor,” Proc. of the Workshop on Multimodal User Authentication, pp.138-143, 2003.
[36] Y. P. Lin, Y. P. Chao, C. C. Lin, and J. H. Chen, “Webcam Mouse Using Face and Eye Tracking in Various Illumination Environments,” Proc. 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, Sep. 1-4, 2005.
[37] M. Soriano, B. Martinkauppi, S. Huovinen, and M. Laaksonen, “Using the Skin Locus to Cope with Changing Illumination Conditions in Color-based Face Tracking,” Proc. IEEE Nordic Signal Processing Symposium, June. 13-15, pp. 383–386, 2000.
[38] Wei-Ming Chen, Hsin-Hung Chiang, Kai-Di Chang, Han-Chieh Chao, Jiann-Liang Chen, “Backup routing firewall mechanism in P2P environment”, Peer-to-Peer Networking and Applications, Vol. 7, Issue 1, pp. 16-23, March 2014.
[39] Wei-Ming Chen, Hsin-Hung Chiang, Han-Chieh Chao, Chao-Hsi Huang, Kai-Ho Chang, “Peer-to-Peer Based Mechanism for Multi-Server System”, Journal of Internet Technology (JIT), Vol. 11, No. 6, pp. 769-776, November 2010.
[40] Apache JMeter, https://jmeter.apache.org/.