MULTI-SCALE DERIVATIVE OF IRREGULAR FUNCTIONS

  • Kim, Tae-Sik (School of Electronical Engineering and Computer Science, Kyungpook National University)
  • Published : 2003.09.01

Abstract

In general, a differential operator can be used as a tool of treating the local properties of given function. However, when the given function is varied with high frequency and has irregular form with non-stationary evolution it may not act its role sufficiently as in case of nowhere differentiable curves. In this paper we introduce a multi-scale derivative as a form of weakened global derivative so that it may explain its semi global diffusion properties as well as local ones for the various irregular diffusion phenomena.

Keywords

References

  1. IEEE Trans. Inform. Theory v.34 Singnal detection in fractional Gaussian noise R.J.Barton;H.V.Poor
  2. SIAM J. Numer. Anal. v.29 Image selective smoothing and edge detection by nonlinear diffusion F.Catte,T.Coll;P.L.Lions;J.M.Morel
  3. Comput. Math. Appl. v.39 Image denoising and segmentation via nonlinear diffusion Y.Chen;B.C.Vemuri;L.Wang
  4. Fractical Geometry K.Falconer
  5. Fractional calculus & applied analysis v.4 no.2 Relations between dimensions and differentiability of curves T.S.Kim;S.Kim
  6. Chaos v.6 Fractional differentiability of nowhere differentiable functions and dimensions K.M.Kolwankar;A.D.Gangal
  7. IEEE Trans. Signal Processing v.46 Method for defining a class of fractional operations P.Kraniauskas;G.Cariolaro;T.Erseghe
  8. Applied Mathematics Letters v.9 The Fundamental solutions for the fractional diffusion-wave equation F.Mainardi
  9. Graphical Models Image Process v.58 Image processing, flows under min/max curvature and mean curvature R.Malladi;J.A.Sethian
  10. Variational Methods in Image Segmentation J.M.Morel;S.Solimini
  11. Chaos, Solutions & Fractals v.4 Singular fractal functions and mesoscopic diffects in mechanics A.Mosolov
  12. Level Set Methods J.A.Sethian
  13. UCLA Math. Dept. CAM Report 96-7 Relation of regularization parameter and scale in total variation based image denoising D.Strong;T.Chan