• DocumentCode
    572938
  • Title

    Application of wavelet de-noising in measurements of the printed line properties

  • Author

    Xu Yan-fang ; Ding Yingkun ; Li Luhai ; Wang Bingzhang

  • Author_Institution
    Sch. of Printing & Packaging Eng., Beijing Inst. of Graphic Commun., Beijing, China
  • fYear
    2012
  • fDate
    24-26 Aug. 2012
  • Firstpage
    867
  • Lastpage
    870
  • Abstract
    In order to realize measurement for digitally printed line´s attributes in accordance with the international standard ISO13660, wavelet de-noising technique is used in pre-processing the CCD digital image with high resolution of hardcopy line prints. First, the line attributes and their measurement indexes are presented. Then, Symltes (SymN) and Coiflets (CoifN) wavelets with hard threshold method are chosen based on theories and experiments. Further, dividing the image signals into two sections, signals of the line itself and the background, and analyzing their de-noising effects with different indexes, the proper wavelet parameters, which are order N in the wavelet functions and decomposition levels, are determined for a high resolution image of a line printed on a ink-jet printer. Experimental results indicate that the most properly parameter N and decomposition levels are 5 and 3 or 4 respectively for SymN, and for CoifN the results are 3 and 3 or 4 respectively. They can satisfy the de-noising requirements of a line CCD image with objectively keeping the line information and the same time smoothing the background information at the most extent.
  • Keywords
    CCD image sensors; ISO standards; image denoising; wavelet transforms; CCD image; CoifN; Coiflets wavelets; SymN; Symltes wavelets; de-noising effects; digitally printed line; international standard IS013660; printed line properties; time smoothing; wavelet de-noising; wavelet de-noising technique; wavelet decomposition; wavelet functions; Barium; Educational institutions; Reflectivity; Very large scale integration; Wavelet transforms; Digital prints; High resolution digital image; Line attributes; Wavelet de-noising;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Information Processing (CSIP), 2012 International Conference on
  • Conference_Location
    Xi´an, Shaanxi
  • Print_ISBN
    978-1-4673-1410-7
  • Type

    conf

  • DOI
    10.1109/CSIP.2012.6308991
  • Filename
    6308991