• DocumentCode
    147291
  • Title

    Retrevial of the original image by artificial neural network

  • Author

    Sankaran, K. Sakthidasan ; Srinithya, G. ; Nagarajan, Vijay

  • Author_Institution
    Electron. & Commun. Dept., Adhiparasakthi Eng. Coll., Melmaruvathur, India
  • fYear
    2014
  • fDate
    3-5 April 2014
  • Firstpage
    1798
  • Lastpage
    1802
  • Abstract
    Restoration along with noise removal is a challenging task in image processing. The image quality is degraded because of the variety of noises. Such noises are removed during restoration in order to obtain the original image as possible. But the removal of noise itself is a major drawback since depicting the type of the noise and the amount of noise is the toughest task. There exist so many filtering and non-filtering techniques for removing the noise. But each of these techniques has several advantages and drawbacks. Filtering techniques involves various filters for restoration and the non filtering techniques are the various algorithms and the artificial neural network methodology used for the restoration. The proposed technique involves using the levenberg marquardt algorithm for restoration. The simulation results shows the usefulness of the proposed method.
  • Keywords
    filtering theory; image denoising; image restoration; image retrieval; neural nets; Levenberg Marquardt algorithm; artificial neural network methodology; filtering technique; image processing; image quality; image restoration; noise removal; nonfiltering techniques; original image retrieval; Filtering; Image edge detection; Image restoration; Neurons; Noise; Prediction algorithms; Training; Artificial neural network; Edge preservation; Levenberg marquardt algorithm; Noise level; restoration;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications and Signal Processing (ICCSP), 2014 International Conference on
  • Conference_Location
    Melmaruvathur
  • Print_ISBN
    978-1-4799-3357-0
  • Type

    conf

  • DOI
    10.1109/ICCSP.2014.6950156
  • Filename
    6950156