• DocumentCode
    2271772
  • Title

    Gaussian Noise Removal of Image on the Local Feature

  • Author

    He, Kun ; Luan, Xin-Cheng ; Li, Chun-Hua ; Liu, Ran

  • Author_Institution
    Comput. Coll., Sichuan Univ., Chengdu
  • Volume
    3
  • fYear
    2008
  • fDate
    20-22 Dec. 2008
  • Firstpage
    867
  • Lastpage
    871
  • Abstract
    The traditional removing algorithm of Gaussian noise can only reduce the effect of noise rather than remove it. Furthermore, the noise points in the image will diffuse after removing. According to the effect of the Gaussian noise on the visual images, this paper introduces an algorithm based on the local feature of the image to eliminate Gaussian noise, and this method overcomes the defects of traditional methods. Firstly, we categorize the location of the pixels into three classes-on the noise point, on the edge, and in the local texture, based on the local continuous smoothing in the image. Secondly, we can extract the edge information and texture of the image by morphology according to the local continuity of the image edge and texture property, then we can accurately locate the noise points of the image. Lastly, we use adaptive neighborhood to eliminate the other noise points. Comparing to the traditional methods, this algorithm can remove the noise better and have satisfying image visual impression.
  • Keywords
    Gaussian noise; adaptive filters; edge detection; feature extraction; image denoising; image texture; smoothing methods; Gaussian noise removal; adaptive filtering; image edge location; image visual impression; local continuous image smoothing; local feature; local image texture; noise point location; pixel location; visual images; Filters; Frequency; Gaussian distribution; Gaussian noise; Histograms; Morphology; Noise level; Noise reduction; Pixel; Smoothing methods; Gaussian Noise; Local neighborhood feature; the edge and texture location;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-0-7695-3497-8
  • Type

    conf

  • DOI
    10.1109/IITA.2008.552
  • Filename
    4740121