• DocumentCode
    3269263
  • Title

    Estimation of PSF for a Shaking Blurred Image Using GA

  • Author

    Shimomukai, Takumi ; Yoshioka, Michifumi ; Yanagimoto, Hidekazu

  • Author_Institution
    Dept. of Comput. Sci. & Intell. Syst., Osaka Prefecture Univ., Sakai, Japan
  • Volume
    2
  • fYear
    2011
  • fDate
    18-21 Dec. 2011
  • Firstpage
    342
  • Lastpage
    345
  • Abstract
    Development of downsized digital cameras causes a lot of undesirable shaking blurred images. In order to restore these images, we need to estimate the PSF (Point Spread Function) from them. It has been proposed to estimate the PSF using cepstrum images. There are the PSF features in the cepstrum images of blurred images. We can estimate the PSF by searching pixels in the cepstrum images. However, cepstrum images also show features of ground truth images which don´t have effects from shaking, so we can´t estimate the PSF accurately. We propose an estimation method of the PSF by using GA.We adopt the cepstrum images as a fitness function,thus the estimation results aren´t affected by other features in cepstrum images.We have just confirmed the effectiveness of our proposed method by using some simulations.
  • Keywords
    cepstral analysis; genetic algorithms; image restoration; optical transfer function; cepstrum image; downsized digital camera; fitness function; genetic algorithm; image restoration; point spread function; shaking blurred image; Cepstrum; Correlation; Educational institutions; Estimation; Image recognition; Image restoration; Image segmentation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Machine Learning and Applications and Workshops (ICMLA), 2011 10th International Conference on
  • Conference_Location
    Honolulu, HI
  • Print_ISBN
    978-1-4577-2134-2
  • Type

    conf

  • DOI
    10.1109/ICMLA.2011.165
  • Filename
    6147701