臺灣博碩士論文加值系統

主動性輪廓模型(Active Contour Model)是由Kass等學者，於1988年所提出，應用於擷取邊界的輪廓。傳統主動性輪廓模型雖可有效的擷取物件邊界，但在實作上的一些限制條件仍然有待克服。此模型最主要的兩大限制在於初始點的設置方式及對於凹面邊界輪廓無法得到有效的擷取效果。傳統模型設置初始點時，必須非常接近邊界，對於物件邊界凹面處的輪廓也因為內部能量的牽引，往往造成凹面處無法取得較好的逼近收斂。
本研究提出了一個新的方法，稱為“分段主動性輪廓模型”(Piece-Wise Active Contour Model)，以克服對於凹面邊界輪廓無法擷取的障礙，在本研究當中，我們利用邊界分佈函數(Edge Distributed Function)的資訊來作為設定初始點的依據，有效的獲得較佳的初始點，用來改善人工設置初始點的繁瑣過程，針對背景單純的影像(如醫學影像)，採取全自動設定初始點；針對背景複雜的自然影像，採取半自動設定初始點，再依據背景複雜程度的不同，以不同方法作修正，以達到簡化設定初始點的效果。對於物件外形為凹面的部分，我們以分段的方式，解決傳統模型在這一方面的缺陷。
根據實驗的結果，對於凹面邊界輪廓，曲線已經可以逼近邊緣且分割出較完整的影像物件，相較於傳統的方法，我們已達到了預期的成果。另外，我們也針對背景複雜程度不同的影像，套用不同設定初始點的方法，使得我們得以簡化設定初始點的繁瑣過程。

Active Contour Model proposed by Kass et al. in 1988 was used for object contour recognition. There are two main drawbacks on traditional active contour algorithm. First, the initial contour must be close to the true boundary. Second, it is difficult to converge a contour to the cavities of an object.
This thesis proposes a new method, called Piece-Wise Active Contour Model, to overcome the drawbacks of the traditional model. In order to improve the efficiency and accuracy of computation, the setting of the initial nodes is based on the information of Edge Distributed Function. This allows initial nodes to attach to object boundary. After the obtained initial points provided the major geometric structures near the edge of an image, the contour is partitioned into several pieces of contours according to its corner information.
Experimental results indicate that the proposed initialization process can be effectively capturing the boundaries. Moreover, Piece-Wise Active Contour Model yields a good convergence at cavities of object.

摘要 ……………………………………………………………………i
Abstract………………………………………………………………iii
致謝 ……………………………………………………………………iv
目錄 ……………………………………………………………………v
圖目錄…………………………………………………………………vii
表目錄 …………………………………………………………………ix
第一章 簡介 ……………………………………………………………1
1.1 研究背景 ……………………………………………………1
1.2 問題描述 …………………………………………………………2
1.3 論文架構 …………………………………………………………3
第二章 相關研究的回顧 ……………………………………………4
2.1主動輪廓模型的相關研究 ………………………………… 4
2.2物件邊緣偵測技術 ………………………………………… 17
第三章 分段主動性輪廓模型………………………………………21
3.1 邊界分佈函數 ……………………………………………………22
3.2 初始點之設置 ……………………………………………………31
3.3 分段主動性輪廓模型 ……………………………………………41
3.3.1 主動性輪廓模型分段的方法 …………………………………41
3.3.2 分段模型能量最佳化 ………………………………42
3.3.3 邊界條件 ………………………………………………………44
第四章 實驗數據與結果……………………………………………46
4.1 背景單純之醫學影像應用結果 …………………………………46
4.2 背景複雜之自然影像應用結果 …………………………………48
4.3 與傳統模型相比較之結果 ………………………………………49
4.3.1 初始點設置方法比較 ……………………………………49
4.3.2 分段方法比較 ……………………………………………52
4.3.3 初始點點數比較 …………………………………………57
第五章 結論與未來發展……………………………………………62
參考文獻 ………………………………………………………………63

[1] M. Kass, A. Witkin and D. Terzopoulos, “Snakes: Active Contour Models,” International Journal of Computer Vision., VOL. 1, pp. 321-331, 1988.
[2] L. D. Cohen, “On Active Contour Models and Balloons,” In CVGIP: Image Understanding, VOL. 53(2), pp. 211-218, MARCH 1991.
[3] C. Xu and J. L. Prince, “Snakes, Shapes, and Gradient Vector Flow,” IEEE Transactions on Image Processing, VOL. 7, NO.3, pp. 359-369, MARCH 1998.
[4] H. Park, T. Schoepflin and Y. Kim, “Active Contour Model with Gradient Directional Information: Directional Snake,” IEEE Transactions on Circuits and Systems for Video Technology, VOL. 11, NO. 2, pp. 252-256, FEBRUARY 2001.
[5] J. Park and J. M. Keller, “Snakes on the Watershed,” IEEE Transactions on Pattern Analysis and Machine Intelligence, VOL. 23, NO. 10, pp. 1201-1205, OCTOBER 2001.
[6] C. Han, T. S. Hatsukami and Jenq-Neng Hwang, “A Fast Mimimal Path Active Contour Model,” IEEE Transactions on Image Processing, VOL. 10, NO 6, pp. 865-873, JUNE 2001.
[7] S. Sun, D. R. Haynor and Y. Kim, ”Semiautomatic Video Object Segmentation Using VSnakes,” IEEE Transactions on Circuits and Systems for Video Technology, VOL. 13, pp. 75-82, 1 January 2003.
[8] Seok-Woo Jang, E. A. El-Kwae and Hyung-Il Choi, “Shaking Snakes Using Color Edge for Contour Extraction,” IEEE ICIP 2002, VOL. 2, pp. II-817 - II-820, 22-25 September 2002.
[9] N. Peterfreund, “The Velocity Snake,” IEEE Proceeding, pp. 70-79, 16 June 1997.
[10] Y. Y. Wong, P. C. Yuen and C. S. Tong, “Segmented Snake for Contour Detection,” Pattern Recognition, VOL. 31, pp. 1669-1679, 1998.
[11] Kap-Ho Seo, W. Kim, C. Oh and Ju-Jang Lee, “Face Detection and Facial Extraction Using Color Snake,” IEEE International Symposium on Industrial Electronics, VOL. 2, pp. 457-462, 8-11 JULY 2002.
[12] T. McInerney and D. Terzopoulos, “Topologically Adaptable Snake,” Int. Conf. on Computer Vision (ICCV’95), pp. 840-845, JUNE 1995.
[13] T. McInerney and D. Terzopoulos, “Topology Adaptive Deformable Surfaces for Medical Image Volume Segmentation,” IEEE Transactions on Medical Imaging, VOL. 18, NO. 10, pp. 840-850, OCTOBER 1999.
[14] F. Leymarie and M. D. Levine, “Tracking Deformable Objects in the Plane Using an Active Contour Model,” IEEE Transactions on Pattern Analysis and Machine Intelligence, VOL. 15, pp. 617-634, 6 June 1993.
[15] H. Jiang and M. S. Drew, “A Predictive Contour Inertia Snake Model for General Video Tracking,” IEEE ICIP 2002, VOL. 3, pp. III-413 -III-416, June 2002.
[16] Dae-Sik Jang and Hyung-I1 Choi, “Active Models for Tracking Moving Objects,” Pattern Recognition, VOL. 33, pp. 1135-1146, July 2000.
[17] Y. Fu, A. T. Erdem and A. M. Tekalp, “Tracking Visible Boundary of Objects Using Occlusion Adaptive Motion Snake,” IEEE Transactions on Image Processing, VOL. 9, NO. 12, pp. 2051-2060, DECEMBER 2000.
[18] W. kim, Sun-Gi Hong and Ju-Jang Lee, “An Active Contour Model Using Image Flow for Tracking a Moving Object,” IEEE International Conference on Intelligent Robots and Systems, VOL. 1, pp. 216-221, 17-21 OCTOBER 1999.
[19] D. Terzopoulos and R. Szeliski, “Tracking with Kalman Snakes,” in Active Vision:3-20, MIT Press,1992.
[20] N. Peterfreund, “Robust Tracking of Position and Velocity with Kalman Snakes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, VOL. 21, pp. 564-569, 6 June 1999.
[21] Wei-Ying Ma and B. S. Manjunath, “EdgeFlow：A Technique for Boundary Detection and Image Segmentation,” IEEE Transactions on Image Processing, VOL. 9, NO. 8, pp. 1375-1388, AUGUST 2000.