Details

Autor(en) / Beteiligte

• Arnéodo, A.
• Decoster, N.
• Kestener, P.
• Roux, S.G.

Titel

A wavelet-based method for multifractal image analysis: From theoretical concepts to experimental applications

Ist Teil von

• Advances in Imaging and Electron Physics, 2003, Vol.126, p.1-92

Ort / Verlag

Elsevier Science & Technology

Erscheinungsjahr

2003

Beschreibungen/Notizen

• This chapter highlights a detailed description of the 2D wavelet transform modulus maxima (WTMM) methodology with some test applications to random monofractal and multifractal self-affine surfaces displaying isotropic as well as anisotropic (with respect to space variables) scale similarity properties. As an illustration of the efficiency and reliability of this method, the chapter explains the main results of its application to experimental 2D data in various domains—namely, geophysics, hydrodynamics, and medicine. For practical purposes, the WTMM representation is emphasized as a very efficient and accurate numerical tool for scanning the singularities of fractal landscapes. The 2D WTMM method is used for many purposes in image processing, including edge detection, pattern recognition, and image denoising. The chapter reviews the outcomes of the statistical analysis of high-resolution LANDSAT satellite images of cloudy scenes. This study brings into light the underlying multiplicative structure of marine stratocumulus clouds. The multifractal properties of the stratocumulus radiance fields are further compared to previous experimental estimates performed on velocity and temperature fluctuations in high Reynolds number turbulence. It has also been indicated that the texture discriminatory power of the 2D WTMM method may lead to significant improvement in computer-assisted diagnosis in digitized mammograms.