On scale-invariant properties in natural images and their simulations

TL;DR

This paper investigates the scale-invariant statistical properties of natural images, demonstrates their presence through power spectrum analysis, and proposes a dynamic model to replicate these properties.

Contribution

It introduces a dynamic model that qualitatively reproduces the scale-invariant power spectrum observed in natural images.

Findings

Power-law decaying power spectrum observed in natural images.

The proposed model qualitatively reproduces the observed spectral slope.

Scale-invariance confirmed numerically for different source types.

Abstract

We study samples of natural images for which a set of statistical characteristics is computed and scale-invariant properties of samples are demonstrated computationally. Computations of the power spectrum are carried out and a power-law decaying power spectrum is observed on samples taken from van Hateren images of natural scenes. We propose a dynamic model to reproduce the observed slope in the power spectrum qualitatively. For two types of sources for this model the behaviour of power spectrum is investigated and scale-invariance confirmed numerically. We then discuss potential applications of scale-invariant properties of natural images.