Information measures in scale-spaces

This correspondence investigates Renyi's generalized entropies under linear and nonlinear scale-space evolutions of images. Scale-spaces are useful co mputer vision concepts for both scale analysis and image restoration. We re gard images as densities and prove monotony and smoothness properties for t he generalized entropies. The scale-space extended generalized entropies ar e applied to global scale selection and size estimations. Finally, we intro duce an entropy-based fingerprint description for textures.