• DocumentCode
    2152702
  • Title

    Characterization of three-dimensional data with multidimensional deformable models based on B-splines in the fourier domain

  • Author

    Berenguer-Vidal, R. ; Verdú-Monedero, R. ; Morales-Sánchez, J. ; Larrey-Ruiz, J. ; Bastida-Jumilla, M.C.

  • Author_Institution
    Dipt. Tecnol. de la Informacion y las Comun., Univ. Politec. Cartagena, Cartagena, Spain
  • fYear
    2011
  • fDate
    22-27 May 2011
  • Firstpage
    709
  • Lastpage
    712
  • Abstract
    This paper presents the formulation and implementation of multidimensional deformable models in the frequency domain. The models are defined by finite element discretization, using B-splines as shape function. The frequency-based formulation allows us to obtain an efficient iterative system in matrix form, suitable for models of any dimension. The aim of this approach is the use of these models for fast segmentation and motion tracking of three-dimensional objects with non rigid motion or deformation from three-dimensional datasets. To show the results of the process, a two dimensional model is applied for the characterization of topographic data.
  • Keywords
    Fourier analysis; frequency-domain analysis; image motion analysis; image segmentation; iterative methods; matrix algebra; object tracking; splines (mathematics); B-splines; Fourier domain; fast image segmentation; frequency domain; iterative system; multidimensional deformable models; shape function; three-dimensional data characterization; three-dimensional datasets; three-dimensional object motion tracking; topographic data characterization; Active contours; Computational modeling; Data models; Deformable models; Mathematical model; Shape; Spline; B-spline; frequency domain; motion tracking; multidimensional deformable model; three-dimensional segmentation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
  • Conference_Location
    Prague
  • ISSN
    1520-6149
  • Print_ISBN
    978-1-4577-0538-0
  • Electronic_ISBN
    1520-6149
  • Type

    conf

  • DOI
    10.1109/ICASSP.2011.5946502
  • Filename
    5946502