James-Stein Type Center Pixel Weights for Non-Local Means Image Denoising

TL;DR

This paper introduces James-Stein type center pixel weights for Non-Local Means image denoising, providing a statistically robust and adaptive method that improves denoising performance across various noise levels.

Contribution

It presents a novel James-Stein based formulation for center pixel weights in NLM and develops an adaptive, locally tuned CPW method for enhanced denoising robustness.

Findings

Higher mean PSNR and SSIM scores

Reduced variance in denoising performance

Improved robustness to noise level variations

Abstract

Non-Local Means (NLM) and variants have been proven to be effective and robust in many image denoising tasks. In this letter, we study the parameter selection problem of center pixel weights (CPW) in NLM. Our key contributions are: 1) we give a novel formulation of the CPW problem from the statistical shrinkage perspective; 2) we introduce the James-Stein type CPWs for NLM; and 3) we propose a new adaptive CPW that is locally tuned for each image pixel. Our experimental results showed that compared to existing CPW solutions, the new proposed CPWs are more robust and effective under various noise levels. In particular, the NLM with the James-Stein type CPWs attain higher means with smaller variances in terms of the peak signal and noise ratio, implying they improve the NLM robustness and make it less sensitive to parameter selection.