本論文提出一即時平行化居家服務型機器人之視覺系統，其中包括了四個主要的獨立子系統。第一個子系統為人臉偵測與追蹤系統，該子系統以自適應的膚色偵測子，平行計算粒子的Condensation濾波器以及類哈爾特徵分類器為基礎架構，並提出了一種簡單快速的人臉運動預測器。第二個子系統為人臉辨識系統，利用嵌入式隱藏馬可夫模型實現，並平行處理訓練與辨識過程。第三個子系統為基於SURF偵測子與描述子的平行式物件偵測系統，其具有尺度與旋轉不變性。第四個子系統進行立體校正與重建，並使用去除失真與極線對齊後的影像，完成立體成像得到三維資訊以進行物件定位。整合四個子系統構成一個快速強健、有效且具可擴充性的平行化整合視覺系統，並使用三顆網路攝影機與多核心筆記型電腦完成實作。本論文的主要貢獻在於提出一種視覺系統的平行化架構。並對在不同未知場景下不同的人與物進行了實驗，實驗結果表明本系統在即時運行時除上述優點外亦具有安全簡單與容易應用等特性。

This thesis presents a real-time parallel vision system for home service robot (HSR). This vision system is set up by four key individual sub-systems. The first one is face detection and tracking sub-system based on adaptive skin detector, condensation filter with parallel computing particles, and Haar-like classifier. And a simple and fast motion predictor is also proposed for face tracking. The second is face recognition system based on embedded HMM with parallel training and recognition procedures. The third is parallel object detection by SURF detector and descriptor, which is scale-invariant and rotation-invariant. The last sub-system is object localization determined by parallel stereo calibration and rectification. Together, these add up to a fast, robust, efficient, and extensible parallel integrated vision system (PIVS), which can run at three webcams and a laptop with multi-core processors. The main contribution of this thesis is the design of parallel structure for a real-time vision system. Finally, real-time experimental results on different humans and objects in different unknown scenes demonstrate that the proposed PIVS is indeed feasible, simple and safe.

[1] G. Bradski and A. Kaehler, Learning OpenCV: Computer vision with the OpenCV library: O'Reilly Media, 2008.
[2] G. Bradski, V. Pisarevsky, and J. Y. Bouguet, “Open source computer vision library,” Intel Corporation, 2001.
[3] OpenCV Wiki. Available: http://opencv.willowgarage.com/wiki/
[4] Intel Corporation. Available: http://www.intel.com/
[5] Willow Garage. Available: http://www.willowgarage.com/
[6] T. Watson. videoInput Library. Available: http://muonics.net/school/spring05/videoInput/
[7] C. Evans, “Notes on the opensurf library,” University of Bristol, Tech. Rep. CSTR-09-001, 2009.
[8] OpenSURF - The Official Home of the Image Processing Library. Available: http://www.chrisevansdev.com/opensurf.html
[9] Microsoft. DirectShow (Windows). Available: http://msdn.microsoft.com/en-us/library/dd375454(VS.85).aspx
[10] Microsoft Corporation. Available: http://www.microsoft.com
[11] OpenMP.org. Available: http://openmp.org/wp/
[12] Intel. Threading Building Blocks.org. Available: http://www.threadingbuildingblocks.org/
[13] Logitech – Get immersed in the digital world with a mouse, keyboard, webcam, headset, Harmony remote, speakers, and more. Available: http://www.logitech.com/en-us/home
[14] HP - Laptops, Desktop, Printers, Servers, and more. Available: http://welcome.hp.com/country/us/en/welcome.html#Product
[15] HP ProBook 6440b specs | Premium Geeks. Available: http://www.premiumgeeks.com/laptop/hp-laptop/hp-probook-6440b-specs/
[16] E. Hjelmås and B. K. Low, “Face Detection: A Survey,” Computer Vision and Image Understanding, vol. 83, pp. 236-274, 2001.
[17] Y. Freund and R. Schapire, “A desicion-theoretic generalization of on-line learning and an application to boosting,” in Proc. Computational Learning Theory, 1995, pp. 23-37.
[18] P. Viola and M. Jones, “Rapid object detection using a boosted cascade of simple features,” in Proc. IEEE CVPR, 2001, pp. I-511-I-518 vol.1.
[19] R. Lienhart and J. Maydt, “An extended set of haar-like features for rapid object detection,” in Proc. IEEE ICIP, 2002, pp. 900-903.
[20] H. Kruppa, M. Castrillón-santana, and B. Schiele, “Fast and robust face finding via local context,” in Proc. IEEE VS-PETS, 2003, pp. 157-164.
[21] M. Castrillón, O. Déniz, C. Guerra, and M. Hernández, “Encara2: Real-time detection of multiple faces at different resolutions in video streams,” Visual Communication and Image Representation, vol. 18, pp. 130-140, 2007.
[22] M. Castrillón-santana, O. Déniz-suárez, L. Antón-canalís, and J. Lorenzo-navarro, “Face and facial feature detection evaluation-performance evaluation of public domain haar detectors for face and facial feature detection,” in Proc. VISAPP 2 INSTICC, 2008, pp. 167-172.
[23] P. Viola and M. Jones, “Robust real-time face detection,” International journal of computer vision, vol. 57, pp. 137-154, 2004.
[24] F. Dadgostar, A. Sarrafzadeh, and S. Overmyer, “Face Tracking Using Mean-Shift Algorithm: A Fuzzy Approach for Boundary Detection,” Affective Computing and Intelligent Interaction, pp. 56-63, 2005.
[25] F. Dadgostar and A. Sarrafzadeh, “An adaptive real-time skin detector based on Hue thresholding: A comparison on two motion tracking methods,” Pattern Recognition Letters, vol. 27, pp. 1342-1352, 2006.
[26] F. Dadgostar and A. Sarrafzadeh, “A Fast Real-Time Skin Detector for Video Sequences,” Image Analysis and Recognition, pp. 804-811, 2005.
[27] A. Koschan and M. Abidi, Digital color image processing: Wiley-Interscience New York, NY, USA, 2008.
[28] R. Kalman, “A new approach to linear filtering and prediction problems,” Journal of Basic Engineering, vol. 82, pp. 35-45, 1960.
[29] K. Daijin and S. Jaewon, Automated Face Analysis: Emerging Technologies and Research: Medical Info Science Reference, 2009.
[30] M. Turk and A. Pentland, “Face recognition using eigenfaces,” in Proc. IEEE CVPR, 1991, pp. 586-591.
[31] W. Zhao, R. Chellappa, and P. J. Phillips, “Subspace linear discriminant analysis for face recognition,” Center for Automation Research, University of Maryland, College Park, Technical Report CAR-TR-914, vol. 137, pp. 138-144, 1999.
[32] P. Belhumeur, J. Hespanha, and D. Kriegman, “Eigenfaces vs. fisherfaces: Recognition using class specific linear projection,” in Proc. Computer Vision-ECCV'96, pp. 43-58, 1996.
[33] T.F. Cootes, G.J. Edwards, and C.J. Taylor, “Active appearance models,” in Proc. Computer Vision-ECCV'98, p. 484-498, 1998.
[34] A. V. Nefian and M. H. Hayes III, “A hidden Markov model-based approach for face detection and recognition,” Georgia Institute of Technology, Atlanta, GA, 1999.
[35] A. V. Nefian and M. H. Hayes, III, “An embedded HMM-based approach for face detection and recognition,” in Proc. IEEE ICASSP, 1999, pp. 3553-3556 vol.6.
[36] F. Samaria and S. Young, “HMM-based architecture for face identification,” Image and vision computing, vol. 12, pp. 537-543, 1994.
[37] S. Theodoridis and K. Koutroumbas, Pattern recognition, 4th ed.: Academic Press, 2009.
[38] A. V. Nefian and M. H. Hayes, III, “Face recognition using an embedded HMM,” in Proc. IEEE AVBPA, 1999, pp. 19-24.
[39] L. Welch, “Hidden markov models and the baum-welch algorithm,” IEEE Information Theory Society Newsletter, vol. 53, pp. 1-10, 2003.
[40] G. David Forney, Jr., “The viterbi algorithm,” Proceedings of the IEEE, vol. 61, pp. 268-278, 1973.
[41] H. Bay, et al., “Speeded-up robust features (SURF),” Computer Vision and Image Understanding, vol. 110, pp. 346-359, 2008.
[42] D. Brown, “Close-range camera calibration,” Photogrammetric Engineering, vol. 37, pp. 855-866, 1971.
[43] J. Fryer and D. Brown, “Lens distortion for close-range photogrammetry,” Photogrammetric Engineering and Remote Sensing, vol. 52, pp. 51-58, 1986.
[44] R. Lenz and D. Fritsch, “Accuracy of videometry with CCD sensors,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 45, pp. 90-110, 1990.
[45] S. Lanser and A. Roboter, Modellbasierte Lokalisation gestutzt auf monokulare Videobilder: Techn. Univ. Munchen, 1997.
[46] S. Lanser, C. Zierl, and R. Beutlhauser, “Multibildkalibrierung einer CCD-Kamera,” in Proc. Mustererkennung, Informatik aktuell, 1995, pp. 481-491.
[47] A. Woods, T. Docherty, and R. Koch, “Image distortions in stereoscopic video systems,” Stereoscopic displays and applications IV, pp. 36-48, 1915.
[48] Z. Zhang, “A flexible new technique for camera calibration,” Pattern Analysis and Machine Intelligence, IEEE Trans. on, vol. 22, pp. 1330-1334, 2000.
[49] J. Bouguet. (2008, Camera calibration toolbox for matlab. Available: http://www.vision.caltech.edu/bouguetj/calib_doc/
[50] Z. Zhang, “Flexible camera calibration by viewing a plane from unknown orientations.,” in Proc. IEEE ICCV, 1999, pp. 666-673.
[51] H. Janne and S. Olli, “A four-step camera calibration procedure with implicit image correction,” in Proc. IEEE CVPR, 1997, pp. 1106-1112.
[52] R. Tsai, “A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses,” IEEE Journal of robotics and Automation, vol. 3, pp. 323-344, 1987.
[53] P. F. Sturm and S. J. Maybank, “On plane-based camera calibration: A general algorithm, singularities, applications,” in Proc. IEEE CVPR, 1999, pp. 432-437.
[54] T. Clarke and J. Fryer, “The development of camera calibration methods and models,” The Photogrammetric Record, vol. 16, pp. 51-66, 1998.
[55] D. Brown, “Decentering distortion of lenses (Lens decentering distortion using thin prism and Conrady models, noting application of precise calibration to analytical photogrammetry),” Photometric Engineering, pp. 444-462, 1965.
[56] A. Conrady, “Decentering lens systems,” Monthly Notices of the Royal Astronomical Society, vol. 79, pp. 384-390, 1919.