• DocumentCode
    256254
  • Title

    Image reconstruction with better edge enhancement using super resolution algorithm

  • Author

    Cheref, Yamina ; Yousfi, D.

  • Author_Institution
    Div. of Earth Obs., Center of Space Tech. - CTS, Arzew, Algeria
  • fYear
    2014
  • fDate
    14-16 April 2014
  • Firstpage
    283
  • Lastpage
    288
  • Abstract
    In this paper, we present super-resolution method that enhance image by combining two algorithms: Iterative algorithm of back-projection (IBP) for removing blur and noise and the algorithm of anisotropic diffusion (AD) which is improved contours. Our technique not only reconstructs a high-resolution image from several overlapping noisy low- resolution images, but also enhances edges and image contrast while suppressing image noise during the restoration process. In order to validate the proposed method and studying its performance, firstly, a synthetic images intentionally degraded are tested, and then an application is performed on satellite images. The obtained results have allowed us to conclude on the validity of our method and its effectiveness to produce a high resolution image and enhancing edges.
  • Keywords
    backpropagation; image denoising; image enhancement; image reconstruction; image resolution; interference suppression; iterative methods; anisotropic diffusion; back-projection iterative algorithm; edge enhancement; image contrast; image noise suppression; image reconstruction; satellite images; super resolution algorithm; Equations; Extraterrestrial measurements; Image edge detection; Image resolution; Image restoration; Kernel; anisotropic diffusion (AD); high-resolution (HR); iterative back projection (IBP); low resolution (LR); super-resolution (SR);
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Multimedia Computing and Systems (ICMCS), 2014 International Conference on
  • Conference_Location
    Marrakech
  • Print_ISBN
    978-1-4799-3823-0
  • Type

    conf

  • DOI
    10.1109/ICMCS.2014.6911234
  • Filename
    6911234