Dense Residual Transformer for Image Denoising
Image denoising is an important low-level computer vision task, which aims to reconstruct a noise-free and high-quality image from a noisy image. With the development of deep learning, convolutional neural network (CNN) has been gradually applied and achieved great success in image denoising, image...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-01-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/11/3/418 |
_version_ | 1797488302816755712 |
---|---|
author | Chao Yao Shuo Jin Meiqin Liu Xiaojuan Ban |
author_facet | Chao Yao Shuo Jin Meiqin Liu Xiaojuan Ban |
author_sort | Chao Yao |
collection | DOAJ |
description | Image denoising is an important low-level computer vision task, which aims to reconstruct a noise-free and high-quality image from a noisy image. With the development of deep learning, convolutional neural network (CNN) has been gradually applied and achieved great success in image denoising, image compression, image enhancement, etc. Recently, Transformer has been a hot technique, which is widely used to tackle computer vision tasks. However, few Transformer-based methods have been proposed for low-level vision tasks. In this paper, we proposed an image denoising network structure based on Transformer, which is named DenSformer. DenSformer consists of three modules, including a preprocessing module, a local-global feature extraction module, and a reconstruction module. Specifically, the local-global feature extraction module consists of several Sformer groups, each of which has several ETransformer layers and a convolution layer, together with a residual connection. These Sformer groups are densely skip-connected to fuse the feature of different layers, and they jointly capture the local and global information from the given noisy images. We conduct our model on comprehensive experiments. In synthetic noise removal, DenSformer outperforms other state-of-the-art methods by up to 0.06–0.28 dB in gray-scale images and 0.57–1.19 dB in color images. In real noise removal, DenSformer can achieve comparable performance, while the number of parameters can be reduced by up to 40%. Experimental results prove that our DenSformer achieves improvement compared to some state-of-the-art methods, both for the synthetic noise data and real noise data, in the objective and subjective evaluations. |
first_indexed | 2024-03-10T00:01:07Z |
format | Article |
id | doaj.art-38fdc433c4a34f3aa9b6a7c81a254182 |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-10T00:01:07Z |
publishDate | 2022-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-38fdc433c4a34f3aa9b6a7c81a2541822023-11-23T16:16:37ZengMDPI AGElectronics2079-92922022-01-0111341810.3390/electronics11030418Dense Residual Transformer for Image DenoisingChao Yao0Shuo Jin1Meiqin Liu2Xiaojuan Ban3School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing 100083, ChinaInstitute of Information Science, Beijing Jiaotong University, Beijing 100044, ChinaInstitute of Information Science, Beijing Jiaotong University, Beijing 100044, ChinaInstitute of Artificial Intelligence, University of Science and Technology Beijing, Beijing 100083, ChinaImage denoising is an important low-level computer vision task, which aims to reconstruct a noise-free and high-quality image from a noisy image. With the development of deep learning, convolutional neural network (CNN) has been gradually applied and achieved great success in image denoising, image compression, image enhancement, etc. Recently, Transformer has been a hot technique, which is widely used to tackle computer vision tasks. However, few Transformer-based methods have been proposed for low-level vision tasks. In this paper, we proposed an image denoising network structure based on Transformer, which is named DenSformer. DenSformer consists of three modules, including a preprocessing module, a local-global feature extraction module, and a reconstruction module. Specifically, the local-global feature extraction module consists of several Sformer groups, each of which has several ETransformer layers and a convolution layer, together with a residual connection. These Sformer groups are densely skip-connected to fuse the feature of different layers, and they jointly capture the local and global information from the given noisy images. We conduct our model on comprehensive experiments. In synthetic noise removal, DenSformer outperforms other state-of-the-art methods by up to 0.06–0.28 dB in gray-scale images and 0.57–1.19 dB in color images. In real noise removal, DenSformer can achieve comparable performance, while the number of parameters can be reduced by up to 40%. Experimental results prove that our DenSformer achieves improvement compared to some state-of-the-art methods, both for the synthetic noise data and real noise data, in the objective and subjective evaluations.https://www.mdpi.com/2079-9292/11/3/418image denoisingresidual skip connectiontransformer |
spellingShingle | Chao Yao Shuo Jin Meiqin Liu Xiaojuan Ban Dense Residual Transformer for Image Denoising Electronics image denoising residual skip connection transformer |
title | Dense Residual Transformer for Image Denoising |
title_full | Dense Residual Transformer for Image Denoising |
title_fullStr | Dense Residual Transformer for Image Denoising |
title_full_unstemmed | Dense Residual Transformer for Image Denoising |
title_short | Dense Residual Transformer for Image Denoising |
title_sort | dense residual transformer for image denoising |
topic | image denoising residual skip connection transformer |
url | https://www.mdpi.com/2079-9292/11/3/418 |
work_keys_str_mv | AT chaoyao denseresidualtransformerforimagedenoising AT shuojin denseresidualtransformerforimagedenoising AT meiqinliu denseresidualtransformerforimagedenoising AT xiaojuanban denseresidualtransformerforimagedenoising |