• DocumentCode
    1503867
  • Title

    Approximation of surfaces in quantitative 3-D reconstructions

  • Author

    Mercer, Robert R. ; Mccauley, Gary M. ; Vel, Satish Anjil

  • Author_Institution
    Duke Univ., Durham, NC, USA
  • Volume
    37
  • Issue
    12
  • fYear
    1990
  • Firstpage
    1136
  • Lastpage
    1146
  • Abstract
    In serial section reconstructions, a series of planar profiles are taken representing curves on the surface of the structure to be reconstructed. For a number of quantitative serial section methods, approximation of a surface is done by the formation of tiles between points of adjacent profiles. As generally proposed, finding this approximation has been difficult due to the inordinately large number of possible solutions resulting from different combinations of tiles between points. Current algorithms have either applied heuristic criteria to force the formation of only one solution, or have searched all acceptable combinations for one that minimizes some cost function. The algorithm presented has been developed to choose the tiling which minimizes the estimated error when the tile approximation of the surface is used in subsequent quantitative algorithms such as the calculation of surface area.
  • Keywords
    computerised picture processing; medical diagnostic computing; applied heuristic criteria; planar profiles; quantitative 3D reconstructions; quantitative algorithms; serial section reconstructions; surface error calculation; Approximation algorithms; Area measurement; Biology computing; Biomedical imaging; Computer errors; Cost function; Helium; Image reconstruction; Surface reconstruction; Tiles; Algorithms; Image Processing, Computer-Assisted; Mathematical Computing; Surface Properties;
  • fLanguage
    English
  • Journal_Title
    Biomedical Engineering, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9294
  • Type

    jour

  • DOI
    10.1109/10.64458
  • Filename
    64458