Deep Differential Convolutional Network for Single Image Super-Resolution
The deep convolutional neural networks and residual networks have shown great success and high-quality reconstruction for single image super-resolution. It is clearly seen that among the best-known super-resolution models, deep learning-based methods demonstrate state-of-the-art performance. In this...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8662556/ |
id |
doaj-65f5f451625e4c3283ea4b900b3c19c5 |
---|---|
record_format |
Article |
spelling |
doaj-65f5f451625e4c3283ea4b900b3c19c52021-03-29T22:14:29ZengIEEEIEEE Access2169-35362019-01-017375553756410.1109/ACCESS.2019.29035288662556Deep Differential Convolutional Network for Single Image Super-ResolutionPeng Liu0https://orcid.org/0000-0002-5362-7769Ying Hong1Yan Liu2Key Laboratory of Information Technology for Autonomous Underwater Vehicles, Institute of Acoustics, Chinese Academy of Science, Beijing, ChinaKey Laboratory of Information Technology for Autonomous Underwater Vehicles, Institute of Acoustics, Chinese Academy of Science, Beijing, ChinaKey Laboratory of Information Technology for Autonomous Underwater Vehicles, Institute of Acoustics, Chinese Academy of Science, Beijing, ChinaThe deep convolutional neural networks and residual networks have shown great success and high-quality reconstruction for single image super-resolution. It is clearly seen that among the best-known super-resolution models, deep learning-based methods demonstrate state-of-the-art performance. In this paper, we propose a deep differential convolutional network (DCN) for single image super-resolution (SRDCN). The proposed DCN is a novel convolutional network, which is composed of convolutional layers, parametric rectified linear units (PReLU), and the identity skip connection. Different from other deep learning-based methods which complete the reconstruction by learning the mapping function between low-resolution and high-resolution images, the proposed algorithm makes changes to the way of reconstruction. In the proposed network, we use DCN to obtain the reconstructed images and the differences between the low-resolution and reconstructed images in the reconstruction process. Then the differences combined with the original low-resolution image and the reconstructed image that from the last DCN are used for final reconstruction. In addition, the loss function is more rationally designed and optimized in this paper. The proposed loss function contains three parts of loss: feature loss, style loss, and mean squared error (MSE) loss. These losses will be used to supervise the structure and content of the reconstructed image. The experimental results prove that the proposed model is superior to many state-of-the-art super-resolution methods in terms of both peak signal-to-noise ratio (PSNR) and structural similarity index metrics (SSIM).https://ieeexplore.ieee.org/document/8662556/Deep convolutional neural networksdifferential convolutional networksingle image super-resolutionpeak signal-to-noise ratiostructure similarity index metrics |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Peng Liu Ying Hong Yan Liu |
spellingShingle |
Peng Liu Ying Hong Yan Liu Deep Differential Convolutional Network for Single Image Super-Resolution IEEE Access Deep convolutional neural networks differential convolutional network single image super-resolution peak signal-to-noise ratio structure similarity index metrics |
author_facet |
Peng Liu Ying Hong Yan Liu |
author_sort |
Peng Liu |
title |
Deep Differential Convolutional Network for Single Image Super-Resolution |
title_short |
Deep Differential Convolutional Network for Single Image Super-Resolution |
title_full |
Deep Differential Convolutional Network for Single Image Super-Resolution |
title_fullStr |
Deep Differential Convolutional Network for Single Image Super-Resolution |
title_full_unstemmed |
Deep Differential Convolutional Network for Single Image Super-Resolution |
title_sort |
deep differential convolutional network for single image super-resolution |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
The deep convolutional neural networks and residual networks have shown great success and high-quality reconstruction for single image super-resolution. It is clearly seen that among the best-known super-resolution models, deep learning-based methods demonstrate state-of-the-art performance. In this paper, we propose a deep differential convolutional network (DCN) for single image super-resolution (SRDCN). The proposed DCN is a novel convolutional network, which is composed of convolutional layers, parametric rectified linear units (PReLU), and the identity skip connection. Different from other deep learning-based methods which complete the reconstruction by learning the mapping function between low-resolution and high-resolution images, the proposed algorithm makes changes to the way of reconstruction. In the proposed network, we use DCN to obtain the reconstructed images and the differences between the low-resolution and reconstructed images in the reconstruction process. Then the differences combined with the original low-resolution image and the reconstructed image that from the last DCN are used for final reconstruction. In addition, the loss function is more rationally designed and optimized in this paper. The proposed loss function contains three parts of loss: feature loss, style loss, and mean squared error (MSE) loss. These losses will be used to supervise the structure and content of the reconstructed image. The experimental results prove that the proposed model is superior to many state-of-the-art super-resolution methods in terms of both peak signal-to-noise ratio (PSNR) and structural similarity index metrics (SSIM). |
topic |
Deep convolutional neural networks differential convolutional network single image super-resolution peak signal-to-noise ratio structure similarity index metrics |
url |
https://ieeexplore.ieee.org/document/8662556/ |
work_keys_str_mv |
AT pengliu deepdifferentialconvolutionalnetworkforsingleimagesuperresolution AT yinghong deepdifferentialconvolutionalnetworkforsingleimagesuperresolution AT yanliu deepdifferentialconvolutionalnetworkforsingleimagesuperresolution |
_version_ |
1724191959923294208 |