從一系列的二維影像重建出其原本的三維模型已經廣泛地應用在各個領域當中，三維模型的呈現方式也有很多種不同的方式。本論文所針對的重建對象為以輪廓線主的二維影像，讓使用者能更具體的看到這一些輪廓線事實上在人類的生活中所代表的真實情況。這些二維的輪廓線可以代表的是人類的器官，例如：腎臟、肝臟．．．等等，也可以是顯微鏡下老鼠腦內的血管，重建這些肉眼無法直接看見的器官成為三維的模型能幫助使用者進一步地去發現出更具有貢獻的研究。

　　在本論文裡設計且實作了一套重建三維物件的系統，每一張以輪廓線為主的二維影像其所包含的東西就是一堆在XY座標上的點的集合，這些點的集合就是輪廓線。最後再將這些點利用OpenGL的方式來建立連結以呈現其原本應有的面貌。

　　利用本系統，可以根據使用者所選擇的輪廓線，並依照使用者的需求來設定連結條件將之建立成三維模型，例如：上下張輪廓線有重疊即建立連結。系統也可以自動依照連結條件的設定來建立出所有可能的三維物件模型，使用者再依其需求來手動修改（新增連結或刪除連結）建立出來的三維物件，以節省使用者的操作時間。

　　除此之外，還提供手動新增輪廓線、編輯輪廓線、刪除輪廓線．．．等等相關操作功能，以滿足使用者對輪廓線的操作。例如：使用者可以將系統切換成重疊顯示的模式，並針對中間那張輪廓線來作平移或是旋轉的操作，系統也會即時地在三維模型的部分作更新。

　　To reconstruct the 3D model from a series of 2D images are now in widespread use in every field. There are also many methods to display the 3D model. The reconstruction targets of this thesis are the contour-based 2D images and it will let the user concretely realize how the contours really exist in our life. The contours can be the organs of human being, for example：the kidney, the liver．．．etc. It can also be the vessels inside the rat brain under the microscope. It was helpful to the user to discover more significant research by reconstructing the invisible objects.

　　In this thesis, we design and implement a 3D reconstruction system. A contour-based image contains a lots of the point sets in the Cartesian coordinate system. This point sets are the contours. The system will reconstruct its original appearance by using the OpenGL method.

　　With this system, users can set up connection conditions for constructing 3D model to meet their needs base on their chosen contours, such as overlap of upper and lower contours could establish connections. More efficiently, the system can also construct every possible 3D models automatically base on the predetermined connection conditions, users can then modify it manually (adding or deleting connections) according to their needs.
　　
　　Furthermore, the system provides additional functions including adding, editing, deleting contours to fulfill users’ needs. For instance, user can translate or rotate the middle slice when the system was operated in the overlap mode, the system would then update the 3D model immediately.

[Li03] Li Tang, Chengke Wu, Tsui, H.T., Shigang Liu, “Algorithm for 3D reconstruction with both visible and missing data”, Electronics Letters, Volume: 39, Issue: 23, Pages:1640-2, 13 Nov. 2003
[Daube01] Daube-Witherspoon, M.E., Matej, S., Karp, J.S., “Assessment of image quality with a fast fully 3D reconstruction algorithm”, Nuclear Science Symposium Conference Record, 2001 IEEE ,Volume: 4, Pages:2238 – 2242, 4 Nov. 2001
[Safont99] Vicent Safont, L., Martinez Marroquin, E., “3D reconstruction of third proximal femur (31-) with active contours”, Image Analysis and Processing, 1999 Proceedings. International Conference, Pages:458 – 463, 27 Sept. 1999
[Ahme97] Ahmed, M., Yamany, S., Hemayed, E., Ahmed, S., Roberts, S., Farag, A., “3D reconstruction of the human jaw from a sequence of images”, Computer Vision and Pattern Recognition, 1997.
[Ande97] Jerry Anderson, John Berg, Michael Regelski, Allen Clark, “ActiveX Programming with Viaual C++ 5.0”, Que corporation, 1997
[Chen96] Jin-Long Chen, Stockman, G.C., “Indexing to 3D model aspects using 2D contour features”, Computer Vision and Pattern Recognition, 1996. Proceedings CVPR '96, 1996 IEEE Computer Society Conference, Page(s): 913 –920, 18 June 1996
[Cohen90] Cohen, L.D., Cohen, I, “A finite element method applied to new active contour models and 3D reconstruction from cross sections”, Computer Vision, 1990. Proceedings, Third International Conference, Pages:587 – 591, 4 Dec. 1990