本研究針對人體的管道影像，預先利用區域成長的方式分割人體的管道影像，並以神經和血管以及腸道影像做測試，加以改進，此外我們也發展一套特徵評估機制，希望在嘗試各種不同的影像特徵時能夠透過評估機制幫助我們找出最佳的分割特徵值及參數。
　　接著我們利用分割出的管道輪廓線，建立管道的3維立體影像，並利用切割工具以及管道分析工具將連續的管道切割成數個分段剖面，最後並搭配表面材質貼圖，呈現管道內部形狀特徵以及管道周圍的器官材質影像。
　　本研究所實作之系統可提供醫療人員在人體管道病變分析上的取得較佳的影像資訊，管道立體影像有別於以往單純的2維影像照影，可取得較佳的形狀輪廓資訊，增加分析與判斷的準確度，此外對於系統的使用者介面操作我們也加以改善，並提供許多工具方便醫療人員在使用操作上更加的流暢。

This research segments images of tubular objects by region growing method. We try to do segmentation for some case, such as a nerve, a blood vessel, and intestinal tract… that to help us to improve this work. Besides, we develop a system for feature evaluation. We try to find out a better feature value to segment images by this evaluation system.

We get the contours of tubular objects by segmentation, and we use the contours to rebuild three-dimensional object, and we use the three-dimensional cutting tools and tube analyzing tools to segment tubular objects to several partition, and collocate surface texture mapping tools to display inside structure of tubular objects and other organs surface texture neighboring tubular objects.

In this study, the system can provide better shape information for pathological analysis of human tubular objects. 3D reconstruction of tubular objects is different in 2D tube images that can provide more inside shape of Tubular objects and to improve the accuracy of pathological analysis. And we try to make our system to more friendly on user interface, and we provide some tools for convenience.

[1]Rafael C. Gonzalez/Richard E. Woods “Digital Image Processing” Second Edition,2002
[2]Lorensen, W. E., Cline, H. E., “Marching cubes: A high resolution 3D surface construction algorithm,” ACM SIGGRAPH Computer Graphics, 1987, v.21 n.4, p.163-169.
[3] Wiki http://zh.wikipedia.org/wiki/File:Standard_deviation_diagram.svg
[4] Wiki
http://zh.wikipedia.org/wiki/%E5%88%87%E6%AF%94%E9%9B%AA%E5%A4%AB%E4%B8%8D%E7%AD%89%E5%BC%8F
[5] Wiki
http://zh.wikipedia.org/wiki/HSL%E5%92%8CHSV%E8%89%B2%E5%BD%A9%E7%A9%BA%E9%97%B4
[6] Tien-Hsiu Tsai,” Image Segmentation Methods Using Evaluation-Based Active Contour Model and Knowledge Source”,2007
[7] N.Otsu,”A threshold selection method from gray level histograms,’’ IEEE Trans.Syst.ManCybern.SMC-9,62–66 1979.”
[8]Wiki
http://en.wikipedia.org/wiki/Coefficient_of_determination
[9]黃哲斌，建立人體器官地圖及數位學習方法，台南，國立成功大學碩士論文，2011
[10] Frank D. Luna, Introduction to 3D Game Programming with DirectX 9.0, WordwarePublishing, 2003.
[11]林纓如，三維物件處理於虛擬手術及模擬的應用，台南，國立成功大學碩士論文，2009
[12] Bright, S., and Laflin, S, Shading of solid voxel models. Computer Graphics Forum 1986, 5,2, 131-138.
[13] Chen, L. S., Herman, G. T., Reynolds, R. A., and Udupa, J. K, Surface shading in the Cuberille environment. IEEE Computer Graphics and Applications, 1985, 5, 12, 33-43.