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

摘要: Publisher Summary 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 anisotropic (with respect space variables) scale similarity properties. As an illustration efficiency reliability this method, explains main results its application experimental data in various domains—namely, geophysics, hydrodynamics, medicine. For practical purposes, WTMM representation is emphasized very efficient accurate numerical tool for scanning singularities fractal landscapes. The method used many purposes image processing, including edge detection, pattern recognition, denoising. reviews outcomes statistical analysis high-resolution LANDSAT satellite images cloudy scenes. study brings into light underlying multiplicative structure marine stratocumulus clouds. properties radiance fields are further compared previous estimates performed on velocity temperature fluctuations high Reynolds number turbulence. It has also been indicated that texture discriminatory power may lead significant improvement computer-assisted diagnosis digitized mammograms.