由於多媒體的應用日益廣泛，以內容為基礎的多媒體視訊應用，像是視訊檢索、視訊辨識、視訊內容分析等，都需要在視訊中將使用者有興趣的物件從影片中切割出來，以達到這些應用的目的，因此視訊物件切割在上述應用中以前處理的身份佔有一個很重要的位置。

半自動視訊物件切割法可分成以區域為基礎（region-based）、和以輪廓為基礎（contour-based）等演算法。以區域為基礎的演算法會有過度切割（over segment）和區塊判斷錯誤的問題，因此在選擇每個區塊的歸屬上甚為 耗時。而以輪廓為基礎的演算法中之蛇形演算法（Snakes），則是利用一些控制點的相對位置的周圍之能量值來決定控制點移向的位置以便得到新的輪廓。但利用能量值來決定輪廓之位置則需要反覆計算而耗費大量的計算時間。為了提高運算速度及其精確度，利用其他特徵值調整輪廓點的位置是一個可行的解決方案。

本論文提出一套使用色彩特徵和輪廓偵測之半自動視訊物件切割法。使用者僅需在欲切割物件上劃一條線以指定此物件，再用相似度比對出的粗糙輪廓和邊界偵測法得出之邊界作交集運算以產生物件初始輪廓。而在後續頁框之輪廓追蹤上，則在前一輪廓上指定一些調整點，而利用其周圍的色彩特徵來區分該邊界點附近的點是否歸屬於此物件，最後將指定之邊界點移到輪廓上適當的位置，並將其他屬於物件之點加以連接即得到精確之物件輪廓之輪廓，並達到視訊物件追蹤之目的。

此一利用色彩特徵和輪廓偵測的半自動視訊物件切割法經實驗後可以發現其切割的精確度比單純使用能量的威特比-蛇形半自動視訊物件切割法高。由於不用反覆求出最佳解，因此在執行時間上也可以大幅度的縮短。

Due to more and more multimedia applications, the need for contend- based multimedia video applica- tions, such as video indexing, video identifica- tion, and video understanding are growing. User- interested objects from the videos need to be segmented to reach the purposes of these applica- tions. Therefore, video object segmentation has important position as pre-processing in these applications.
Many semi-automatic video object segmentation methods have been studied, such as region-based methods, contour-based methods etc. The region- based methods need to waste time in deciding every block, because of the problems about over-segment and region decision errors. In contour-based methods, the Snakes method is used the energy functions to get the new contour by using the energy around the corresponding positions of the control points to decide the new positions of con- trol points. But using energy function to decide the new contour wastes a huge amount of computation time on getting the best solution using energy minimization repeatedly. In order to improve the speed and accuracy, it is a good choice to adjust the positions of contour points by using other features.
In this paper we propose a semi-automatic video object segmentation method using color fea- ture and contour detection. Users only need to draw a line on user-interested objects to assign this object, and then generate initial object contour by computing the intersection operation of coarse- contour resulted from comparing similarities and edges obtained form edge detection method. For contour tracking, first some edge points are assigned on previous contour, and then using the color features of the surrounding points to decide whether the points near the edge point belongs to object or not. Next, the edge points are adjusted to the appropriate position. Finally, get the fine contour by connecting the other points belongs to object.
Experiment results show that the precision of segmentation of this semi-automatic video object segmentation method using color feature and contour detection is better than VSnakes semi-automatic video object segmentation method. Without computing the best solution repeatedly so it is possible to save lots of execution time.

[1] J.Liu, D.Przewonzy, and S.Pastoor, “Layered representation of scenes based on multiview image analysis,” IEEE Trans. Circuits Syst. Video Techonl., vol.10, pp.518-529, June 2000.
[2] NGrammalidis and M.G. Strintzis, “Disparity and occlusion estimation in multi ocular systems and their coding for the communication of multi view image sequences,” IEEE Trans. Circuits Syst. Video Techonl., vol.8, pp.328-344, June 1998.
[3] A.Marugame, A.Yamada, and M.Ohta, “Focused object extraction with multiple cameras” IEEE Trans. Circuits Syst. Video Techonl., vol.10,pp.530-540, June 2000.
[4] D.Chai and K.N.Ngan, “Face segmentation using skin-color map in videophone applications,” IEEE Trans. Circuits Syst. Video Techonl., vol.28, pp.551-564, June 1999.
[5] H.Li, “Segmentation of facial area for videophone application” Electron.Lett., vol.28, pp. 1915-1916, Sept 1992.
[6] S.Cochran and G. Medioni, “3-D surface description from binocular stereo,” IEEE Trans. Pattern Anal. Machine Intell., vol.12, pp.981-994, Oct.1992.
[7] E.Izquierdo, “Disparity/segmentation analysis: Matching with an adaptive window and depth-driven segmentation,” IEEE Trans. Circuits Syst. Video Techonl., vol.9, pp.589-607, June 1999.
[8] W. Hoff and N. Ahuja, “Surfaces from stereo: Integrating feature matching, disparity estimation and contour detection,” IEEE Trans. Pattern Anal. Machine Intell., vol.11, pp121-136, Feb.1989.
[9] J.Weng, N. Ahuja, and T.S. Huang, “Matching two perspective views,” IEEE Trans. Pattern Anal. Machine Intell., vol.14, pp.806-825, Aug.1992.
[10] S.L. Gazit and G. Mendioni, “Multi-scale contour matching in a motion sequence,” in Proc. DARPA Image Understanding Workshop, Palo Alto, CA, pp.934-943, 1989.
[11] M. Kirby and L. Sirovich, “Application of the Karhunen-Loève procedure for the characterization of human faces,” IEEE Trans. Pattern Anal. Machine Intell., vol.12, pp.103-108, Jan.1990.
[12] B. Moghaddam and A. Pentland, “Probabilistic visual learning for object detection,” in Proc. 5th Int. Conf. Computer Vision, Cambridge, MA, pp.786– 793, 1995.
[13] K.-K. Sung and T. Poggio, “Example-based learning for view- based human face detection,” IEEE Trans. Pattern Anal. Machine Intell., vol.20, pp.39–51, Jan. 1998.
[14] M.-H. Yang, N. Ahuja, and D. Kriegman, “Face detection using a mixture of factor analyzers,” in Proc. IEEE Int. Conf. Image Processing,1999, pp.612– 614.
[15] S.-H. Lin, S. Y. Kung, and L.-J. Lin, “Face recognition/detection by probabilistic decision-based neural network,” IEEE Trans. Neural Network, vol.8, pp.114–132, Jan. 1997.
[16] H. A. Rowley, S. Baluja, and T. Kanade, “Neural network-based face detection,” IEEE Trans. Pattern Anal. Machine Intell., vol. 20, pp.23–38, Jan. 1998.
[17] J. Ng and S. Gong, “Multiview face detection and pose estimation using a composite vector machine across the view sphere,” in Proc. IEEE Int. Workshop in Recognition, Analysis and Tracking of Faces and Gestures in Real-Time Systems, Corfu, Greece, pp. 14–21, Sept. 1999.
[18] A. J. Colmenarez and T. S. Huang, “Face detection with information-based maximum discrimination,” in Proc. IEEE Conf. Pattern Recognition, pp. 782–787,1997.
[19] M. S. Lew and N. Huijsmans, “Information theory and face detection,” in Proc. IEEE Int. Conf. Image Processing, vol. 3, pp. 601–605, 1996.
[20] H. Wang and S.-F. Chang, “A highly efficient system for automatic face region detection in MPEG video,” IEEE Trans. Circuits Syst. Video Technol., vol. 7, pp.615–628, Aug. 1997.
[21] H. Yang and N. Ahuja, “Detection human faces in color images,” in Proc. IEEE ICIP, Chicago, IL, pp. 127–130, 1998.
[22] Sun, S.J., Haynor, D.R., Kim, Y.M., “Semiautomatic video object segmentation using Vsnakes,” IEEE Trans. Circ. Syst. Video Technol. 13 (1), pp.75–82, 2003.
[23] Luo, H.T., Eleftheriadis, A., “An interactive authoring system for video object segmentation and annotation,” Signal Process.: Image Commun. 17 (7), pp. 559–572, 2002.
[24] Lim, J., Cho, H.K., Beom Ra, J.,. An improved video object tracking algorithm based on motion re-estimation. Proc. IEEE ICIP 1, pp.339–342, 2000.
[25] Kim, Y.R., Kim, J.H., Kim, Y., Ko, S.J., ”Semiautomatic segmentation using spatio-temporal gradual region merging for MPEG-4,” IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences E86-A(10), 2526–2534, 2003.
[26] Kim, M., Jeon, J.G., Kwak, J.S., Lee, M.H., Ahn, C., ”Moving object segmentation in video sequence by userinteraction and automatic object tracking,” Image Vis.Comput. 19 (5), pp.245–260, 2001.
[27] Guo, J., Kim, J.W., Kuo, C.-C.J., “An interactive object segmentation system for MPEG video,” Proc. IEEE ICIP 2, pp.140–144, 1999.
[28] Zhi Liu, Jie Yang, and Ning Song Peng, “Semi-automatic video object segmentation using seeded region merging and bidirectional projection,” Pattern Recognition Letters, v 26, n 5,pp.653-662,2005.
[29] Vincent, L., Soille, P., Watersheds in digital spaces: an efficient algorithm based on immersion simulations. IEEE Trans. Pattern Anal. Mach. Intell. pp.583–598. 1991.
[30] Adams, R., Bischof, L., Seeded region growing. IEEE Trans. Pattern Anal. Machine Intell., pp.641–647,1994.
[31] Patras Ioannis, Hendriks, Emile A, Lagendijk, Reginald L. “Semi-automatic object-based video segmentation with labeling of color segments” Signal Processing: Image Communication, vol.18, n1, pp.51-65, January, 2003
[32] Luo, Huitao; Eleftheriadis, Alexandros “Model-based segmentation and tracking of head-and-shoulder video objects for real time multimedia services”, IEEE Transactions on Multimedia, v5, n3, September, 2003,
[33] M. Kass, A. Witkin, and D. Terzopoulos, “Snakes: Active contour models,” Int. J. Comput. Vis., vol.1, pp.321–331, 1988.
[34] A. J. Viterbi, “Error bounds for convolutional codes and an asymptotically optimum decoding algorithm,” IEEE Trans. Inform. Theory, vol.IT-13, pp. 260–269. 1967.
[35] J Canny, “A computational approach to edge detection” IEEE Transactions on Pattern Analysis and Machine Intelligence Volume 8, Issue 6 Pages: 679 – 698, 1986
[36] Cheng H.D., Jiang X.H., Sun Y., Wang J, “Color image segmentation: Advances and prospects.” Pattern Recognition, v 34, n 12, pp.2259-2281,2001
[37] Hariharakrishnan, K.,and Schonfeld, D., ”Fast object tracking using adaptive block matching” IEEE Transactions on Multimedia, Volume 7, Issue 5, pp. 853 – 859, 2005
[38] Munchurl Kim, Jae Gark Choi, Daehee Kim, Hyung Lee, Myoung Ho Lee, Chieteuk Ahn, Yo-Sung Ho, “A VOP generation tool: automatic segmentation of moving objects in image sequences based on spatial-temporal information,” IEEE Trans. Circuits Syst. Video Techonl., Vol.9, Issue 8, pp.1216–1226, Dec. 1999