• DocumentCode
    2228930
  • Title

    Image Recovery by Recursive Spline Smoothing and Application to Recognition of Compound Structures

  • Author

    Grebennikov, Alexandre

  • Author_Institution
    Fac. of Phys.-Math. Sci., Merited Autonomous Univ. of Puebla, Puebla, Mexico
  • fYear
    2012
  • fDate
    16-18 April 2012
  • Firstpage
    25
  • Lastpage
    29
  • Abstract
    In this paper we propose approach for special data post-processing, based on Recursive Spline Smoothing method, adapted to image reconstruction. The description of the special MATLAB program package is presented. Constructed algorithms and programs are applied to compound structures reconstruction. Necessity of reconstruction of compound structures appears in different applications. As a rule, tomography images are used for this purpose. However, obtained result usually contains significant visible noise. To eliminate such noise specific approach is applied for structures, which have piecewise constant characteristics with known values. It consists in Recursive Spline Smoothing and includes also projection of the pre-reconstructed data to the known set of values with respect to the absolute or relative criterion.
  • Keywords
    computerised tomography; image recognition; image reconstruction; mathematics computing; piecewise constant techniques; splines (mathematics); MATLAB program package; compound structure recognition; compound structure reconstruction; data post-processing; image reconstruction; image recovery; piecewise constant characteristics; recursive spline smoothing method; tomography images; Approximation methods; Compounds; Image reconstruction; Mathematical model; Smoothing methods; Spline; Tomography; compounds; computed tomography; images; reconstruction algorithms;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Technology: New Generations (ITNG), 2012 Ninth International Conference on
  • Conference_Location
    Las Vegas, NV
  • Print_ISBN
    978-1-4673-0798-7
  • Type

    conf

  • DOI
    10.1109/ITNG.2012.12
  • Filename
    6209142