Image Denoising Using Nonlocal Regularized Deep Image Prior
Deep neural networks have shown great potential in various low-level vision tasks, leading to several state-of-the-art image denoising techniques. Training a deep neural network in a supervised fashion usually requires the collection of a great number of examples and the consumption of a significant...
Guardado en:
Autores principales: | , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
MDPI AG
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/f1a59cb80852453da2d0d92f316142c0 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:f1a59cb80852453da2d0d92f316142c0 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:f1a59cb80852453da2d0d92f316142c02021-11-25T19:06:51ZImage Denoising Using Nonlocal Regularized Deep Image Prior10.3390/sym131121142073-8994https://doaj.org/article/f1a59cb80852453da2d0d92f316142c02021-11-01T00:00:00Zhttps://www.mdpi.com/2073-8994/13/11/2114https://doaj.org/toc/2073-8994Deep neural networks have shown great potential in various low-level vision tasks, leading to several state-of-the-art image denoising techniques. Training a deep neural network in a supervised fashion usually requires the collection of a great number of examples and the consumption of a significant amount of time. However, the collection of training samples is very difficult for some application scenarios, such as the full-sampled data of magnetic resonance imaging and the data of satellite remote sensing imaging. In this paper, we overcome the problem of a lack of training data by using an unsupervised deep-learning-based method. Specifically, we propose a deep-learning-based method based on the deep image prior (DIP) method, which only requires a noisy image as training data, without any clean data. It infers the natural images with random inputs and the corrupted observation with the help of performing correction via a convolutional network. We improve the original DIP method as follows: Firstly, the original optimization objective function is modified by adding nonlocal regularizers, consisting of a spatial filter and a frequency domain filter, to promote the gradient sparsity of the solution. Secondly, we solve the optimization problem with the alternating direction method of multipliers (ADMM) framework, resulting in two separate optimization problems, including a symmetric U-Net training step and a plug-and-play proximal denoising step. As such, the proposed method exploits the powerful denoising ability of both deep neural networks and nonlocal regularizations. Experiments validate the effectiveness of leveraging a combination of DIP and nonlocal regularizers, and demonstrate the superior performance of the proposed method both quantitatively and visually compared with the original DIP method.Zhonghua XieLingjun LiuZhongliang LuoJianfeng HuangMDPI AGarticledeep neural networksdeep image prior (DIP)nonlocal regularizerplug-and-playalternating direction method of multipliers (ADMM)MathematicsQA1-939ENSymmetry, Vol 13, Iss 2114, p 2114 (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
deep neural networks deep image prior (DIP) nonlocal regularizer plug-and-play alternating direction method of multipliers (ADMM) Mathematics QA1-939 |
spellingShingle |
deep neural networks deep image prior (DIP) nonlocal regularizer plug-and-play alternating direction method of multipliers (ADMM) Mathematics QA1-939 Zhonghua Xie Lingjun Liu Zhongliang Luo Jianfeng Huang Image Denoising Using Nonlocal Regularized Deep Image Prior |
description |
Deep neural networks have shown great potential in various low-level vision tasks, leading to several state-of-the-art image denoising techniques. Training a deep neural network in a supervised fashion usually requires the collection of a great number of examples and the consumption of a significant amount of time. However, the collection of training samples is very difficult for some application scenarios, such as the full-sampled data of magnetic resonance imaging and the data of satellite remote sensing imaging. In this paper, we overcome the problem of a lack of training data by using an unsupervised deep-learning-based method. Specifically, we propose a deep-learning-based method based on the deep image prior (DIP) method, which only requires a noisy image as training data, without any clean data. It infers the natural images with random inputs and the corrupted observation with the help of performing correction via a convolutional network. We improve the original DIP method as follows: Firstly, the original optimization objective function is modified by adding nonlocal regularizers, consisting of a spatial filter and a frequency domain filter, to promote the gradient sparsity of the solution. Secondly, we solve the optimization problem with the alternating direction method of multipliers (ADMM) framework, resulting in two separate optimization problems, including a symmetric U-Net training step and a plug-and-play proximal denoising step. As such, the proposed method exploits the powerful denoising ability of both deep neural networks and nonlocal regularizations. Experiments validate the effectiveness of leveraging a combination of DIP and nonlocal regularizers, and demonstrate the superior performance of the proposed method both quantitatively and visually compared with the original DIP method. |
format |
article |
author |
Zhonghua Xie Lingjun Liu Zhongliang Luo Jianfeng Huang |
author_facet |
Zhonghua Xie Lingjun Liu Zhongliang Luo Jianfeng Huang |
author_sort |
Zhonghua Xie |
title |
Image Denoising Using Nonlocal Regularized Deep Image Prior |
title_short |
Image Denoising Using Nonlocal Regularized Deep Image Prior |
title_full |
Image Denoising Using Nonlocal Regularized Deep Image Prior |
title_fullStr |
Image Denoising Using Nonlocal Regularized Deep Image Prior |
title_full_unstemmed |
Image Denoising Using Nonlocal Regularized Deep Image Prior |
title_sort |
image denoising using nonlocal regularized deep image prior |
publisher |
MDPI AG |
publishDate |
2021 |
url |
https://doaj.org/article/f1a59cb80852453da2d0d92f316142c0 |
work_keys_str_mv |
AT zhonghuaxie imagedenoisingusingnonlocalregularizeddeepimageprior AT lingjunliu imagedenoisingusingnonlocalregularizeddeepimageprior AT zhongliangluo imagedenoisingusingnonlocalregularizeddeepimageprior AT jianfenghuang imagedenoisingusingnonlocalregularizeddeepimageprior |
_version_ |
1718410315660525568 |