Wavelet and entropy analysis combination to evaluate diffusion and correlation behaviors

Diffusion and correlation effects are two principal phenomena which have been studied for years and several visualization techniques have been proposed to help scientists to understand them. The analysis of these phenomena will help to extract important information from data sets. To understand these problems we combine wavelet and entropy analysis to evaluate the evolution of these behaviors through scale and time. We present image case studies to show several different kinds of behaviors of these effects. Some of them are fallible cases and not reliable, as the images do not show the desired information. We calculate entropy of smooth and detail coefficient sets, generated by wavelet transform of these sample images in each scale, to obtain measures that allow us to evaluate these behaviors according to the organization complexity. These measures can provide an indication about the quality of the rendered images.