• DocumentCode
    1641169
  • Title

    A study of adaptive meshing in FEM-CSI for microwave tomography

  • Author

    Zakaria, Amer ; LoVetri, Joe

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of Manitoba, Winnipeg, MB, Canada
  • fYear
    2010
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    The use of finite element method (FEM) to discretize the differential operator in the contrast source inversion (CSI) offers several advantages over conventional integral-equation based CSI algorithms. One of the advantages of FEM-CSI is the ability to adaptively control the density of the triangular mesh within the inversion domain to increase the quality of the reconstructed image and to reduce the computational burden of the inversion algorithm. Here we present an example that exploits this advantage of FEM-CSI, providing a qualitative and quantitative analysis using several inversion results. The example selected is for an unbounded-region microwave tomography configuration under the 2D transverse magnetic (TM) approximation.
  • Keywords
    approximation theory; image reconstruction; integral equations; mesh generation; tomography; 2D transverse magnetic approximation; FEM-CSI; adaptive density control; adaptive meshing; contrast source inversion; finite element method; image reconstruciton; integral-equation based CSI algorithms; inversion algorithm; microwave tomography; qualitative analysis; quantitative analysis; triangular mesh; unbounded-region microwave tomography; Finite element methods; Image reconstruction; Imaging; Matrix decomposition; Microwave theory and techniques; Permittivity; Sparse matrices;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Antenna Technology and Applied Electromagnetics & the American Electromagnetics Conference (ANTEM-AMEREM), 2010 14th International Symposium on
  • Conference_Location
    Ottawa, ON
  • Print_ISBN
    978-1-4244-5049-7
  • Electronic_ISBN
    978-1-4244-5050-3
  • Type

    conf

  • DOI
    10.1109/ANTEM.2010.5552529
  • Filename
    5552529