A Study on the Delaunay Triangulation Reconstruction using the Modified EVS

수정된 EVS를 이용한 Delaunay 삼각형 재구성에 관한 연구

  • Kwon E.C. (Department of Biomedical Engineering, College of Health Science, Yonsei University) ;
  • Shin D.K. (Department of Biomedical Engineering, College of Health Science, Yonsei University) ;
  • Kim D.Y. (Department of Biomedical Engineering, College of Health Science, Yonsei University)
  • 김동윤 (연세대학교 보건과학대학 의용전자공학과) ;
  • 신동규 (연세대학교 보건과학대학 의용전자공학과) ;
  • 김동윤 (연세대학교 보건과학대학 의용전자공학과)
  • Published : 2000.06.01

Abstract

The Delaunay triangulation. which is one of the surface rendering methods. have been evaluated as a good algorithm which can solve the geometrical connection problems and make high quality images. However this method also have the problem that is the 'non-solid' connection between slices. In this paper, we proposed a new method for the Delaunay triangulation for the surface rendering from 2D medical images in the PC environment. The proposed method was designed to eliminate 'non-solid' tetrahedra. which had no correspondence to the geometric closeness, and used elimination algorithm with modified External Voronoi Skeleton path. This method enabled us to eliminate 'non-solid' tetrahedra without affecting other regions. Thus we could effectively reconstruct the complex shaped objects which were compactly connected with tetrahedra.

표면에 의한 3차원 재구성 방법 중 Delaunay 삼각형 분할 방법은 기존의 기하학적 연결성 문제를 해결하고 영상의 질도 높은 것으로 평가받고 있다. 그러나 이 방법은 윤곽선의 위치에 따른 비 입체적 연결성 문제를 갖고 있다. 본 논문에서는 PC환경에서 이러한 문제를 개선한 Delaunay 삼각형 분할 재구성 방법을 제안하였다. 제안된 방법은 기존의 방법에서 나타나는 '비 입체(non-solid)‘ 사면체의 효과적인 처리를 위하여 수정된 External Voronoi Skeleton(EVS)을 이용한 제거 알고리즘을 사용하였다. 이를 적용한 결과. 원 객체 형태에 영향을 주지 않으면서 '비 입체' 사면체만이 제거됨을 확인할 수 있었다. 그리고 재구성된 영상의 표면이 사면체로 촘촘하게 연결되므로 복잡한 형태를 갖는 입체의 표면을 효과적으로 재구성할 수 있었다.

Keywords

References

  1. IEEE Computer Graphics & Applications v.11 no.1 Computer Graphics in Medicine M. Rodes
  2. 석사학위논문 PC기반의 3차원 의료영상 재구성 시스템의 교속화 설계 배수현
  3. NEUROSUR GERY v.11 no.1 Three Dimensional Computer Reconstruction of Brain Lesions from Surface Contours Provided by CT : A Prospectus S. Batnitzky(et al.)
  4. Report 2105. INRIA Three dimensional modeling of human organs and its application to diagnosis and surgical planning B. Geiger
  5. Annals of New York Academy of Science v.12 The analysis of cell image J.M.S. Prewitt;M.L. Mendelsohn
  6. The Computer Journal v.24 no.2 Computing the n-dimensional Delaunay tessellation with application to Noronoi polytopes D. F. Watson
  7. Computer Vision, Graphics, and Image Processing v.44 Shape Reconstruction from Plannar Cross Sections J. D. Boissonnat
  8. Numerical Method for Engineers D. V. Griffiths;I. M. Smith
  9. 3D Computer Graphics A. Watt
  10. Computer Graphics v.21 no.4 Marching Cubes: A High Resolution 3D Surface Construction Algorithm W. E. Lorensen;H. E. Cline
  11. IEEE 1996 International Conference on Image Processing Marching Triangles: Range Image Fusion for Complex Object Modelling Hilton A.;Stoddart A.J.;Illingworth J.;Windeatt T.
  12. Submitted to IEEE Computer Graphics and Applications Marching Triangles: Surface Based Implicit surface Polygonisation Hilton A.;Illingworth J.