• DocumentCode
    2983705
  • Title

    Simultaneous Textured Image Decomposition and Discrimination Preserving Large-Scale Structure

  • Author

    Shahidi, R. ; Moloney, C.

  • Author_Institution
    Fac. of Eng. & Appl. Sci., St. John´´s, Nfld.
  • fYear
    2006
  • fDate
    Aug. 2006
  • Firstpage
    343
  • Lastpage
    348
  • Abstract
    In their pioneering paper on image decomposition, Vese and Osher (2003) used the auxiliary texture components g1 and g2 for the purposes of texture discrimination. This was done sequentially - first decomposition was performed, after which either |g 1| or |g2| was used for the discrimination phase, which utilized active contours without edges (T. Chan and L. Vese, 1999). With the new simultaneous discrimination and decomposition scheme introduced in the present paper, fewer iterations are required for both the discrimination and decomposition processes, with improved quality of results. In accomplishing this parallel solution of the two problems, we also introduce an active contour model which is robust to noise and which relies both on region and local edge information
  • Keywords
    image texture; auxiliary texture components; discrimination preserving large-scale structure; information technology; local edge information; texture discrimination; textured image decomposition; Active contours; Active noise reduction; Humans; Image decomposition; Image denoising; Image segmentation; Information technology; Large-scale systems; Noise robustness; Signal processing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing and Information Technology, 2006 IEEE International Symposium on
  • Conference_Location
    Vancouver, BC
  • Print_ISBN
    0-7803-9753-3
  • Electronic_ISBN
    0-7803-9754-1
  • Type

    conf

  • DOI
    10.1109/ISSPIT.2006.270823
  • Filename
    4042265