• DocumentCode
    2094516
  • Title

    An Isophote-Oriented Image Interpolation Method

  • Author

    Yi, Zhan ; Minghui, Wang ; Meng, Li

  • Author_Institution
    Coll. of Comput., Sichuan Univ., Chengdu, China
  • Volume
    1
  • fYear
    2008
  • fDate
    20-22 Dec. 2008
  • Firstpage
    723
  • Lastpage
    726
  • Abstract
    This paper presents an isophote-oriented image interpolation method that attempts to produce smooth reconstructions of the image¿s level curves while still maintaining image fidelity. The proposed method employs an interpolation kernel that is obtained by looking for the explicit solution of a nonlinear PDE (partial differential equation). The PDE preserve the local continuation of isophotes (curves of constant intensity). By doing so, the curvature of the interpolated isophotes is reduced, and, thus, zigzagging artifacts are largely suppressed. Moreover, the interpolation techniques can preserve edge sharpness. In addition, the proposed methods can be used irrespective of the magnification factor (MF) and easily implemented due to their simple structure. Experiments show that images interpolated using the proposed method are both visually and quantitatively improved.
  • Keywords
    image reconstruction; interpolation; nonlinear differential equations; partial differential equations; edge sharpness; image fidelity; image level curves; image smooth reconstruction; isophote-oriented image interpolation; magnification factor; nonlinear partial differential equation; Computer science; Digital images; Educational institutions; Image reconstruction; Interpolation; Kernel; Mathematics; PSNR; Paper technology; Partial differential equations; image interpolation; image reconstruction; isophote;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Computational Technology, 2008. ISCSCT '08. International Symposium on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-3746-7
  • Type

    conf

  • DOI
    10.1109/ISCSCT.2008.277
  • Filename
    4731528