S\'eparation en composantes structures, textures et bruit d'une image, apport de l'utilisation des contourlettes

TL;DR

This paper introduces an improved image decomposition method for noisy images by replacing wavelet transforms with contourlet transforms, leading to better geometric approximation and reduced artifacts.

Contribution

It presents a novel approach using contourlet transforms for image decomposition, enhancing the separation of structures, textures, and noise compared to wavelet-based methods.

Findings

Contourlet transform reduces artifacts in image decomposition.

The proposed iterative algorithm effectively separates image components.

Improved geometric approximation in noisy textured images.

Abstract

In this paper, we propose to improve image decomposition algorithms in the case of noisy images. In \cite{gilles1,aujoluvw}, the authors propose to separate structures, textures and noise from an image. Unfortunately, the use of separable wavelets shows some artefacts. In this paper, we propose to replace the wavelet transform by the contourlet transform which better approximate geometry in images. For that, we define contourlet spaces and their associated norms. Then, we get an iterative algorithm which we test on two noisy textured images.