• DocumentCode
    2168102
  • Title

    Combined Nonlinear Iterative Algorithms for Retrieving the Complex Wave Field

  • Author

    Zheng, Shi-Ling ; Xue, Bin-Dang ; Jiang, Zhi-Guo

  • Author_Institution
    Sch. of Astronaut., Beijing Univ. of Aeronaut. & Astronaut., Beijing, China
  • fYear
    2009
  • fDate
    17-19 Oct. 2009
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    Phase retrieval problem has attracted many interests for its wide application in many domains. In this paper we focus on the topic of retrieving the full complex wave field of the object plane from three or more diffraction planes. The iterative Fresnel algorithm always suffers from slow convergence and stagnation problems, while the generalized hybrid-input-output (GHIO) algorithm can effectively accelerate convergence and break stagnation of this kind. Therefore, we combined the two methods into a more efficient algorithm. Each step, the iterative Fresnel algorithm using three or more diffraction intensities is firstly performed to jointly estimate the intensity and the phase of the object plane. Then the GHIO algorithm is adopted to generate a new complex amplitude for the next iteration. It is not necessary to search optimal step size as the conjugate gradient method and can also avoid stagnation, so it is very efficient. This is proved by simulations.
  • Keywords
    conjugate gradient methods; image retrieval; iterative methods; combined nonlinear iterative algorithms; complex wave field; conjugate gradient method; generalized hybrid-input-output algorithm; iterative Fresnel algorithm; phase retrieval problem; Convergence; Extraterrestrial measurements; Gradient methods; Iterative algorithms; Optical diffraction; Optical microscopy; Optical propagation; Optical scattering; Optical sensors; Phase estimation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image and Signal Processing, 2009. CISP '09. 2nd International Congress on
  • Conference_Location
    Tianjin
  • Print_ISBN
    978-1-4244-4129-7
  • Electronic_ISBN
    978-1-4244-4131-0
  • Type

    conf

  • DOI
    10.1109/CISP.2009.5304567
  • Filename
    5304567