Stein Block Thresholding For Image Denoising

TL;DR

This paper analyzes the theoretical minimax properties of Stein block thresholding for image denoising across various dimensions, demonstrating near-optimal rates and practical effectiveness compared to state-of-the-art methods.

Contribution

It establishes the minimax rates of Stein block thresholding over decomposition spaces and provides a practical, fast denoising procedure with theoretical and empirical validation.

Findings

Achieves near-minimax rates in image denoising

Outperforms BLS-GSM in practical tests

Provides a theoretical framework for threshold parameter selection

Abstract

In this paper, we investigate the minimax properties of Stein block thresholding in any dimension $d$ with a particular emphasis on $d=2$. Towards this goal, we consider a frame coefficient space over which minimaxity is proved. The choice of this space is inspired by the characterization provided in \cite{BorupNielsen} of family of smoothness spaces on $\mathbb{R}^d$, a subclass of so-called decomposition spaces \cite{Feichtinger}. These smoothness spaces cover the classical case of Besov spaces, as well as smoothness spaces corresponding to curvelet-type constructions. Our main theoretical result investigates the minimax rates over these decomposition spaces, and shows that our block estimator can achieve the optimal minimax rate, or is at least nearly-minimax (up to a $\log$ factor) in the least favorable situation. Another contribution is that the minimax rates given here are stated for a general noise sequence model in the transform coefficient domain beyond the usual i.i.d. Gaussian case. The choice of the threshold parameter is theoretically discussed and its optimal value is stated for some noise models such as the (non-necessarily i.i.d.) Gaussian case. We provide a simple, fast and a practical procedure. We also report a comprehensive simulation study to support our theoretical findings. The practical performance of our Stein block denoising compares very favorably to the BLS-GSM state-of-the art denoising algorithm on a large set of test images. A toolbox is made available for download on the Internet to reproduce the results discussed in this paper.