Threshold Selection for Total Variation Denoising

TL;DR

This paper introduces an efficient two-step adaptive threshold selection method for total variation denoising, improving segmentation accuracy and reducing computational costs in image and signal processing.

Contribution

It proposes a novel threshold selection procedure based on large deviation theory, offering an alternative to costly cross-validation methods for TV denoising.

Findings

Effective in denoising 1D and 2D signals

Achieves exact segmentation under certain conditions

Reduces computational cost compared to traditional methods

Abstract

Total variation (TV) denoising is a nonparametric smoothing method that has good properties for preserving sharp edges and contours in objects with spatial structures like natural images. The estimate is sparse in the sense that TV reconstruction leads to a piecewise constant function with a small number of jumps. A threshold parameter controls the number of jumps and the quality of the estimation. In practice, this threshold is often selected by minimizing a goodness-of-fit criterion like cross-validation, which can be costly as it requires solving the high-dimensional and non-differentiable TV optimization problem many times. We propose instead a two step adaptive procedure via a connection to large deviation of stochastic processes. We also give conditions under which TV denoising achieves exact segmentation. We then apply our procedure to denoise a collection of 1D and 2D test signals verifying the effectiveness of our approach in practice.