• DocumentCode
    2290487
  • Title

    Connected skeletons from 3D distance transforms

  • Author

    Capson, David W. ; Fung, Andrew Chi-san

  • Author_Institution
    Dept. of Electr. & Comput. Eng., McMaster Univ., Hamilton, Ont., Canada
  • fYear
    1998
  • fDate
    5-7 Apr 1998
  • Firstpage
    174
  • Lastpage
    179
  • Abstract
    An algorithm for generating connected skeletons of binary regions in volumetric images is presented. The method is based on a 3D distance transformation which may be computed in two passes through the input image voxels. We introduce the concept of rings and satellites for the identification of 3D saddle points in the distance transformed image which are used to ensure connectivity in the skeleton
  • Keywords
    feature extraction; identification; image segmentation; transforms; 3D distance transform; 3D saddle points identification; algorithm; binary regions; connected skeletons; distance transformed image; input image voxels; rings; satellites; volumetric images; Bandwidth; Computed tomography; Costs; Data mining; Laser modes; Magnetic resonance imaging; Memory; Ring lasers; Satellites; Skeleton;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Analysis and Interpretation, 1998 IEEE Southwest Symposium on
  • Conference_Location
    Tucson, AZ
  • Print_ISBN
    0-7803-4876-1
  • Type

    conf

  • DOI
    10.1109/IAI.1998.666881
  • Filename
    666881