臺灣博碩士論文加值系統

一般的影像切割技術應用在超音波影像上並無法有很好的效果，主要是因為超音波影像含有斑點, 雜訊及組織紋理。在這篇論文中，我們將合併紋理分析技術及主動尋求輪廓模型在超音波影像上對乳房腫瘤做切割。我們所提出的切割方法包括了兩大部分，分別是取得最初輪廓和輪廓變形階段。在取得最初輪廓部分，我們提出了數種方法來找出好的最初輪廓，使得找出的最初輪廓仍維持腫瘤的形狀且會很街近腫瘤的邊線並存在於腫瘤的內部。在輪廓變形階段，最初輪廓會被拉向腫瘤的邊線位置。我們利用了紋理資訊來定義外力，使得輪廓不會陷入雜訊和組織紋理所造成的錯誤位置中。最後，我們將證明我們所提出的方法應用在超音波影像的乳房腫瘤切割上有令人滿意的效果。

Due to the speckles, noise, and tissue-related textures, most techniques for segmenting ultrasound images do not perform well. In this paper, we combine texture analysis schemes and the discrete active contour model to segment the breast tumors in the ultrasound images. There are two main steps in the proposed segmentation method, the initial contour finding and the snake deformation process. In the initial contour finding, we propose several methods to get a proper initial contour, which maintains the tumor shape and is close to the tumor boundary and inside the tumor. In the deformation process, the contour is pulled toward the boundary of the tumor. We utilize the texture feature to define the external force, so that the snake will not trap into the false position caused by tissues-related textures or noise. Finally, we will show that the proposed techniques have rather good performance and lead to satisfactory result in segmenting the breast tumors in the ultrasound images.

摘 要i
ABSTRACTii
ACKNOWLEDGEMENTSiv
TABLE OF CONTENTSv
LIST OF FIGURESvii
LIST OF TABLESix
CHAPTER 1 INTRODUCTION1
CHAPTER 2 INITIAL CONTOUR FINDING4
2.1Statistical Analysis Schemes4
2.1.1 Standard Deviation Analysis4
2.1.2 Covariance Texture Parameter6
2.2PCA for Speckle Detection and Gaussian Blurring7
2.3Watershed Segmentation9
2.4Stick Detection10
CHAPTER 3 THE CONTOUR DEFORMATION USING SNAKE12
3.1Discrete Active Contour Model12
3.2External Force Using Texture Information13
CHAPTER 4 EXPERIMENT RESULTS AND DISCUSSION16
CHAPTER 5 CONCLUSION20
REFERENCES21

[1]J. Canny, “A computational approach to edge detection,” IEEE Trans. Pattern Anal. Machine. Intell., vol. 8, no. 6, pp. 679-698, Nov. 1986.
[2]T. Pavlidis, Algorithms for graphics and image processing, Rockville, Maryland: Computer Science Press, 1982.
[3]T. Pavlidis and S.L. Horowitz, “Segmentation of plane curves,” IEEE Trans. Comput., vol. 27, no. 8, pp. 860-870, Aug. 1974.
[4]M. Kass, A. Witkin, and D. Terzopoulos, “Snakes: Active Contour models,” Int. J. Comput, Vision, pp. 321-331, 1988.
[5]C. Vieren and F. Cabestaing, “Catching moving objects with snakes for motion tracking,” Pattern Recognition Letters, vol. 16, no.7, pp. 679-685, July 1995.
[6]F. Leymarie and M.D. Levine, “Tracking deformable objects in the plane using an active contour model,” IEEE Trans. Pattern Anal. Machine Intell., vol. 15, no. 6, pp. 617-634, June 1993.
[7]A.K. Jain, Y. Zhong, and S. Lzkshmanan, “Object matching using deformable templates,” IEEE Trans. Pattern Anal. Machine Intell., vol. 18, no. 3, pp. 267-278, Mar. 1996.
[8]L.D. Cohen and I. Cohen, “Finite-element methods for active contour models and balloons for 2D and 3D images,” IEEE Trans. Pattern Anal. Machine Intell., vol. 15, no. 11, pp. 1131-1147. Nov. 1993.
[9]P.C. Yuen, Y.Y. Wong, and C.S Tong , “Contour detection using enhanced snakes algorithm,” Electron. Lett. vol. 32, no. 3, pp. 202-204, Feb. 1996.
[10]C.M. Chen, H.S. Henry, and Y.C. Lin, “An early vision-based snake model for ultrasound image segmentation,” Ultrasound in Med. & Biol., vol. 26, no. 2, pp. 273-285, 2000.
[11]S. Lobregt and M. A. Viergever, “A discrete dynamic contour model,” IEEE Trans. Med. Imag., vol. 14, no. 1, pp. 12—24, Mar. 1995.
[12]R.C. Gonzales and P. Wintz, Digital Image Processing. 2nd Ed., Reading, Massachusetts: Addison-Wesley, 1987.
[13]H. Peckinpaugh, “An improved method for computing gray-level cooccurrence matrix based texture measures,” CVGIP Graphical Models and Image Processing, vol. 53, pp. 574-580, 1991.
[14]C. Roux, J. L. Coatrieux, Contemporary perspectives in three-dimensional biomedical imaging. Amsterdan, Netherland: IOS Press, 1997, p 141.
[15]A. Theresa, F. Jochen, J. Brian and L. Paul, “Automated three-dimensional US frame positioning computed from elevational speckle decorrelation,” Radiology, vol. 209, pp. 575—582, 1998.
[16]C. Choi, M. Wirth, and A. Jennings, “Segmentation: Artificial life and watershed transform approach,” in IEE 6th International Conference on Image Processing and Its Applications, 1997, vol. 1, no. 443, pp. 371-375.
[17]K. Haris, S. N. Efstratiadis, and N. Maglaveras, “Watershed-based image segmentation with fast region merging,” in Proceedings of the IEEE International Conference on Image Processing, 1998, vol. 3, pp. 338-342.
[18]L. Vincent and P. Soille, “Watersheds in digital spaces: an efficient algorithm based on immersion simulations,” IEEE Trans. Pattern Anal. Machine Intell., vol. 13, no. 6, pp. 583—598, June 1991.
[19]K. Haris and S.N. Efstratiadis, “Hybrid image segmentation using watersheds and fast region merging” IEEE Trans. Image Processing, vol. 7, no. 12, pp. 1684-1699, Dec. 1998.
[20]N. Richard and R.N. Czerwinski, “Detection of lines and boundaries in speckle images-application to medical ultrasound,” IEEE Trans. Med. Imag., vol. 18, no. 2, pp. 126-136, Feb. 1999.