近年來，智慧型監控系統一直是熱門的研究項目，而這些研究成果，也多被應用於特殊場所的安全監控系統或特殊路段的交通控管系統當中。在這些相關的研究中，往往依照監控環境與應用目的的不同，系統各環節使用了不盡相同的處理方法，而最終的考量，大抵都是為了達到目標物的偵測與追蹤兩個目的。
在大部分智慧型監控系統的應用與研究，多是針對地面上的行人與車輛，在本篇論文中，我們將這類監控系統的應用推展至空中，希望能夠加強空防系統的效能，因此我們研究了一套針對高空飛行移動物體的偵測追蹤系統。
在偵測與追蹤的過程方面，首先我們利用連續影像間強度的差異來偵測是否畫面中存在飛行移動物體，並配合一些處理，以濾除室外環境中由於光線或其他偶發事件所造成的雜訊干擾，然後再利用影像分割常用的分水嶺分割法，對目標物存在的部分影像區域進行影像分割，以取得目標物的影像資料，如此完成整個偵測流程的處理。之後的追蹤階段，我們利用卡曼濾波器，預測下一個時間點的目標物大略位置座標，再繼續利用分水嶺分割法對預測的位置區域進行影像分割以取得精確的目標物資訊，如此遞迴運作，以達到持續追蹤的目的。
我們以實際拍攝的高空飛行物影像片段來進行系統的實驗，實驗結果也展示出我們的系統在偵測與追蹤高空飛行物方面有相當的正確率與可行性。

Recently, abundant researches relating to unsupervised surveillance systems have been presented which can be applied to security systems and traffic control systems. The composing elements and procedures of these systems are different due to the differences in monitoring environments and application purposes. However, the ultimate consideration is the successful detection and tracking of targets.
In most intelligent surveillance systems, the targets that they handle are mainly pedestrians walking and vehicles driving in the ground. In this thesis, we plan to extend the system further to the sky to strengthen the air defense systems by developing a high altitude flying object detection and tracking system.
In the proposed system, we firstly detect whether flying objects appear in the surveillance field according to the intensity difference of the pixels in contiguous frames. In this step, some processes need to be performed to remove noises caused by lighting variations or other accidental events. Then, watershed transformation is adopted to segment target images. Next, useful information of targets are extracted from the segmented areas where the targets locate. In the tracking phase, Kalman filter is first employed to predict the possible locations of targets. Then, watershed transformation is applied to the predicting positions to segment the regions. Precise target information can be extracted from the segmented images. Perform the steps repeatedly, we can accomplish the goal of continuous tracking.
Experiments were conducted on several image sequences with small targets (aircrafts). Experimental results reveal the feasibility and validity of our proposed system in detecting and tracking high-altitude aircrafts.

[1] N. Peterfreund, “Robust Tracking of Position and Velocity With Kalman Snakes”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, no. 6, June, pp. 564-569, 1999.
[2] N. Paragios and R. Deriche,”Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 3, March, pp. 266-280, 2000.
[3] O. Masoud and N. P. Papanikolopoulos, “A Novel Method for Tracking and Counting Pedestrians in Real-Time Using a Single Camera”, IEEE Transactions on Vehicular Technology, vol. 50, no. 5, September, pp. 1267-1278, 2001.
[4] D. Murray and A. Basu, “Motion Tracking with an Active Camera”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16, no. 5, May, pp. 449-459, 1994.
[5] D. Davies, P. Palmer, and M. Mirmehdi, “Detection and Tracking of Very Small Low Contrast Objects”, British Machine Vision Conference, September, pp. 599-608, 1998.
[6] M. Greiffenhagen, D. Comaniciu, H. Niemann, V. Ramesh, “Design, Analysis, and Engineering of Video Monitoring System: An Approach and a Case Study ”, Proc. of the IEEE, vol. 89, no. 10, October, pp. 1498-1517, 2001.
[7] R. T. Collins, A. J. Lipton, T. Kanade, H. Fujiyoshi, D. Duggins, Y. Tsin, D. Tolliver, N. Enomoto, O. Hasegawa, P. Burt and L. Wixson, ”A System for Video Surveillance and Monitoring”, Tech. Rep., The Robotics Institute, Carnegie Mellon University, 2000. CMU-RI-TR-00-12.
[8] J. Black, T. Ellis, “Multi Camera Image Tracking”, Proceedings 2nd IEEE Int. Workshop on PETS, Kauai, Hawaii, USA, December 9, 2001.
[9] S. Beucher and C. Lantuejoul, “Use of Watersheds in Contour Detection”, International Workshop on Image proc.: Real-Time Eage and Motion Detection / Estimation, Rennes, France, September, pp. 17-21, 1979.
[10] C. R. Jung, and J. Scharcanski, “Robust Watershed Segmentation Using the Wavelet Transform”, Proc. of the XV Brazilian Symposium on Computer Graphics and Image Processing, pp. 1530-1834, 2002.
[11] M. Fisher, and R. Aldridge, “Hierarchical Image Segmentation Using A Watershed Scale-Space Tree”, Image Processing and Its Applications, Conference Publication, no. 465, pp. 522-526, 1999.
[12] J. Park and J. M. Keller, “Snake on Watershed”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 10, October, pp. 1201-1205, 2001.
[13] G. Welch, and G. Bishop,”High Performance Wide-Area Optical Tracking”, Massachusetts Institute of Technology, vol. 10, no. 1, February, pp. 1-21, 2001.
[14] P. A. Ffrench, J. R. Zeidler, and W. H. Ku, ”Enhanced Detectability of Small Objects in Correlated Clutter Using An Improved 2-d Adaptive Lattice Algorithm.”, IEEE Transactions on Image Processing, vol. 3, no. 6, pp. 383 – 397, 1997.
[15] S. Beucher, “The Watershed Transformation Applied to Image Segmentations”, Proc. Pfefferkorn Conf. on Signal and Image Processing in Microscopy and Microanalysis, pp. 99 – 314, September, 1991.
[16] P.D. Smet, R. Luis, and V. P. M. Pires,”Implementation and Analysis of An Optimized Rainfalling Watershed Algorithm”, Proc. Science and Technology Conference: Image and Video Communications and Processings, January, 2000.
[17] S. E. Hernandez, and K. E. Barner, ”Joint Region Merging Criteria for Watershed-Based Image Segmentation.”, Proc. IEEE International Conference on Image Processing, vol. 2, pp. 108 – 111, 2000.
[18] T. Blaffert, S. Dippel, M. Stahl, and R. Wiemker,”The Laplace Integral for A Watershed Segmentation.”, Proc. 2000 International Conference on Image Processing, vol. 3, pp. 444 – 447, 2000.