監控系統已經廣泛地應用在門禁管制、交通監測與犯罪蒐證等方面。但多數系統均以監控廣大視野為目的，無法顧及特定目標物的影像解析度，在實際應用上有其限制。本論文提出一套包含靜態與動態(PTZ)攝影機之多攝影機主動式監視系統以及其攝影機校正方法，使監視系統能同時監控廣大場景並取得目標物高解析度的影像。
本論文重點主要在於靜態攝影機與動態攝影機之間的溝通與對應，以及其中的攝影機參數校正。利用場景中可測的特徵點資訊，對靜態攝影機以及PTZ攝影機進行攝影機參數的校正。系統運作時，靜態攝影機會在實驗場景內進行監控，一旦偵測到有人物進入監視範圍，便根據攝影機校正的結果，計算出人物位於場景中的位置。透過座標轉換，指揮PTZ攝影機轉到適當的角度，並擷取目標的影像。
完成初步的追蹤之後利用轉換FPI的方式，截取膚色的資訊，在PTZ攝影機的視野中進行人臉偵測，判斷人臉是否位於影像中央。並且以人臉位置與影像中心的偏移量，進行PTZ攝影機校準，將PTZ攝影機更準確的對準人臉，最後盡可能的對人臉進行倍率放大，取得高解析度的人臉影像。最後提出以目標物的靜態攝影機影像資訊與PTZ攝影機角度資訊，計算包含高度資訊的目標物實際世界座標的方法，使目標物的資訊更加豐富。
經實驗得知，本論文所提出的攝影機校正方法，可有效應用於以多台攝影機為基礎的監視系統上。透過攝影機對應性，監視系統可以同時監控廣大的場景，並擷取目標物高解析度的人臉影像。即使初步的轉動有所偏差，仍能藉由人臉偵測的資訊，將動態攝影機校準至正確的角度。最後利用目標物與攝影機之間的關係，計算出目標物實際的高度與位置。

Surveillance systems have been widely used in visitor control, traffic monitor and many other applications nowadays. Most of them provides functions such as data recording and simple information analyses. However, blurred images were usually captured and difficult to recognize due to a wide surveillance field. In order to resolve these problems, this thesis proposes an active surveillance which is system composed of a static camera and a PTZ (pan-tilt-zoom) camera with the corresponding cameras calibration methods. The system can monitor a wide field and obtain clear images of the objects.
The geometric relationship between static and PTZ cameras were first settled. Some significant points within the camera scene were chosen as the feature points. The information of these feature points were then used to calibrate both the static camera and PTZ camera. After cameras have been calibrated, the objects entering the surveillance scene will be detected and tracked. Then the system will steer the PTZ camera to get the clear image of the tracked object.
The image of the PTZ camera is then transformed into flesh probability image (FPI) for face detection. The image offset between the human face and the image center will be used to tune the PTZ camera to get the highest quality image of human face. Experimental result show that clear image of human face can be captured by the system even when the initial rotation of the PTZ camera is far away from the object.

[1] S. J. Maybank and O.D. Faugeras, "A theory of self-calibration of a moving camera," The International Journal of Computer Vision, Vol. 8, No. 2, Aug, 1992,pp.123-152.
[2] Q.-T. Luong, "Matrice Fondamentale et Calibration Visuelle sur l'Environment-Vers une plus grande autonomie des system robotiques," PhD thesis, Universite de Paris-Sud, Centre d'Orsay, Dec, 1992.
[3] Kuan-Wen Chen, Yi-Ping Hung, Yong-Sheng Chen, “On Calibrating a Camera Network using Parabolic Trajectories of a Bouncing Ball”, Proceedings of the Second Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance, Beijing, China, Oct. 2005.
[4] Z.-Y. Zhang, "Camera Calibration with One-Dimensional Objects," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 26, No. 7, Jul, 2004, pp.892-899.
[5] P. Sturm and S. Maybank, "On Plane-based Camera Calibration: A General Algorithm, Singularities, Applications," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognitions, Fort Collins, Colorado, USA, Jun, 1999. IEEE Computer Society Press. pp.432-437.
[6] Z.-Y. Zhang, "A Flexible New Technique for Camera Calibration," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 11, Nov, 2000, pp.1330-1334.
[7] Chu-Song Chen, Chi-Kuo Yu, Yi-Ping Hung, “New Calibration-free Approach for Augmented Reality Based on Parameterized Cuboid Structure,” Proceedings of the IEEE International Conference on Computer Vision, Corfu, Greece, 1999, pp.30-37.
[8] H. L. Chou and W. H. Tsai, "A new approach to robot location by house corners," Pattern Recognition, Vol. 19, No. 6, 1986, pp.439-451.
[9] Kai-Tai Song and Jen-Chao Tai, “Dynamic Calibration of Pan-Tilt-Zoom Cameras for Traffic Monitoring,” IEEE Trans. on Systems, Man and Cybernetics, Part B , Vol. 36, No. 5, pp. 1091-1103, 2006.
[10] I-Hsien Chen, Sheng-Jyh Wang, “Efficient Vision-Based Calibration for Visual Surveillance Systems with Multiple PTZ Cameras”, IEEE 2006.
[11] Everts, I., Sebe, N. and Jones, G.A., “Cooperative Object Tracking with Multiple PTZ Cameras”, IEEE 14th International Conference on Image Analysis and Processing, 2007.
[12] Anton Obukhov, Konstantin Strelnikov and Dmitriy Vatolin "Fully Automatic PTZ Camera Calibration Method". Proc. of Graphicon'2008, pp. 122-127, Moscow, Russia, June 2008
[13] 楊上毅、孫永年，“人臉追蹤與辨識之自動化視訊監控系統”，國立成功大學資訊工程研究所碩士論文，2008。
[14] R.L. Hsu, M. Abdel-Mottaleb, A.K. Jain, “Face detection in color images,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 24, pp. 696-706, 2002.
[15] C. Garcia and G. Tziritas, “Face detection using quantized skin color regions merging and wavelet packet analysis,” IEEE Trans. Multimedia, Vol. 1 No. 3, pp. 264-277, 1999.
[16] W. H. Press, W. T. Vetterling, S. A. Teukolsky, B. P. Flannery, “Numerical Recipes in C 2nd Edition,” Cambridge University Press, Cambridge, England, 1992.
[17] H. Stern and B. Efros, “Adaptive color space switching for tracking under varying illumination,” Image and Vision Computing, Vol. 23, pp. 353-364, 2005.