Image denoising with less artefacts: Novel non-linear filtering on fast patch reorderings
Image and Video Processing
2020-02-04 v1 Signal Processing
Abstract
Leading denoising methods such as 3D block matching (BM3D) are patch-based. However, they can suffer from frequency domain artefacts and require to specify explicit noise models. We present a patch-based method that avoids these drawbacks. It combines a simple and fast patch reordering with a non-linear smoothing. The smoothing rewards both patch and pixel similarities in a multiplicative way. We perform experiments on real world images with additive white Gaussian noise (AWGN), and on electron microscopy data with a more general additive noise model. Our filter outperforms BM3D in 77% of the experiments, with improvements of up to 29% with respect to the mean squared error.
Cite
@article{arxiv.2002.00638,
title = {Image denoising with less artefacts: Novel non-linear filtering on fast patch reorderings},
author = {Kireeti Bodduna and Joachim Weickert},
journal= {arXiv preprint arXiv:2002.00638},
year = {2020}
}