Fast CNN enhancement using channel attention and residual networks for image super-resolution

Research Article
Open access

Fast CNN enhancement using channel attention and residual networks for image super-resolution

Haoning Qu 1 , Yifeng Ruan 2 , Zerui Wan 3 , Ming Zhu 4
  • 1 College of Arts and Science, New York University, New York, NY, 10012, USA    
  • 2 Alfred Lerner College of Business & Economics, University of Delaware, Newark, DE, 19711, USA    
  • 3 Faculty of Applied Science & Engineering, University of Toronto, Toronto, ON, M5S 1A4，Canada    
  • 4 New College, University of Toronto, Toronto, ON, M5S 1A1, Canada    
  • *corresponding author
ACE Vol.4
ISSN (Print): 2755-273X
ISSN (Online): 2755-2721
ISBN (Print): 978-1-915371-55-3
ISBN (Online): 978-1-915371-56-0

Abstract

Single image super-resolution (SISR) refers to the process of reconstructing a high-resolution (HR) image from a low-resolution (LR) input image. Deep learning super-resolution algorithms have widely been used to solve SISR tasks. However, the demanding computation cost and memory occupation incurred through training the deep learning models has been hindering its real-world application. In this paper, we rebuild FSRCNN and apply it to solve SISR tasks. Firstly, we change the original training dataset to RealSR, a larger dataset consisting of real-world images. Secondly, channel attention and residual blocks have been applied to the mapping layers and important parameters including learning rate and optimizer have been reset. Thirdly, we change the cost function from loss to loss and replace the activation function from parametric rectified linear unit (PReLU) to exponential linear unit (ELU), to verify the discrepancies between different loss functions and activation functions. Finally, we compare the rebuilt models with the official FSRCNN based on the Peak signal-to-noise ratio (PSNR) and the structural similarity index measure (SSIM) on three common test datasets. The original model achieves better performance on all the test datasets across different scale factors while the rebuilt models show better generalization capability. Our analyses illustrate that residual blocks can slightly promote model performance while different loss functions and activation functions do not generate an evident impact on the rebuilt model.

Keywords:

channel attention, fast convolutional neural networks, residual block, super-resolution

Qu,H.;Ruan,Y.;Wan,Z.;Zhu,M. (2023). Fast CNN enhancement using channel attention and residual networks for image super-resolution. Applied and Computational Engineering,4,134-142.
Export citation

References

[1]. Chunwei T Ruibin Z Zhihao W Yong X Wangmeng Z Chen C and Chia-Wen L 2020 Lightweight image super-resolution with enhanced CNN arXiv eess.IV(2020)2007.04344v3

[2]. Chao D, Chen C L, Kaiming H, and Xiaoou T Image super-resolution using deep convolutional networks arXiv cs.CV (2015)1501.00092v3

[3]. Jin Y Shigesumi K and Takio K Fast and accurate image super resolution by deep CNN with skip connection and network in network arXiv cs.CV (2017) 1707.05425v7

[4]. Jiwon K Jung K L and Kyoung M L Deeply-recursive convolutional network for image super-resolution arXiv cs.CV (2016)1501.04491v2

[5]. Christian L Lucas T Ferenc H Jose C Andrew C Alejandro A Andrew A Alykhan T Johannes T Zehan W and Wenzhe S Photo-realistic single image super-resolution using a generative adversarial network arXiv cs.CV (2017)1609.04802v5

[6]. Zhisheng L Juncheng L Hong L Chaoyan H Linlin Z and Tieyong Z Transformer for single image super-resolution arXiv cs.CV (2022)2108.11084v3

[7]. Chao D Chen C L and Xiaoou T Accelerating the super-resolution convolutional neural network arXiv cs.CV (2016)1608.00367v1

[8]. Zhisheng L Juncheng L Hong L Chaoyan H Linlin Z and Tieyong Z Transformer for single image super-resolution arXiv cs.CV (2022) 2108.11084v3

[9]. Jianrui C Hui Z Hongwei Y Zisheng C and Lei Z Toward real-world single image super-resolution: a new benchmark and a new model arXiv cs.CV (2019) 1904.00523

[10]. Jie H Li S Samuel Al Gang S and Enhua W Squeeze-and-excitation networks arXiv cs.CV (2019) 1709.01507v4

[11]. Kaiming H Xiangyu Z Shaoqing R and Jian S Deep residual learning for image recognition arXiv cs.CV (2015)1512.03385v1

[12]. Jiali W Zhengchun L Ian F Won C Rajkumar K and V. Rao K Fast and accurate learned multiresolution dynamical downscaling for precipitation arXiv cs.LG(2021) 2101.06813v1

[13]. Hang Z Orazio G Iuri F and Jan K Loss functions for image restoration with neural networks arXiv cs.LG(2021) 1511.08861v3

[14]. Alain H and Djemel Z Image quality metrics: PSNR vs. SSIM 2010 International Conference on Pattern Recognition

[15]. Honggang C Xiaohai H Linbo Q Yuanyuan W Chao R and Ce Z Real-world single image super-resolution: a brief review arXiv eess.IV(2021 2103.02368v1

[16]. Yang J Wright J Huang T S and Ma Y Image super resolution via sparse representation arXiv IEEE Transactions on Image Processing, vol. 19, no. 11, pp. 2861-2873 (2010)

[17]. Djork-Arne ́ C Thomas U and Sepp H Fast and accurate deep network learning by exponential linear units (ELUS) arXiv cs.LG(2016) 1511.07289v5

[18]. Xuezhe M Apollo: an adaptive parameter-wise diagonal quasi-newton method for nonconvex stochastic optimization arXiv cs.LG (2016)2009.13586v6

[19]. Bevilacqua M Roumy A Guillemot C and Morel M.L.A. Low-complexity single-image super-resolution based on nonnegative neighbor embedding arXiv BMVC. (2012)

[20]. Zeyde R Elad M and Protter M On single image scale-up using sparse-representations arXiv Curves and Surfaces (2012) 711–730

[21]. Martin D Fowlkes C Tal D and Malik J A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics arXiv ICCV Volume 2 (2001) 416–423

Cite this article

Qu,H.;Ruan,Y.;Wan,Z.;Zhu,M. (2023). Fast CNN enhancement using channel attention and residual networks for image super-resolution. Applied and Computational Engineering,4,134-142.

Data availability

The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.

Disclaimer/Publisher's Note

The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of EWA Publishing and/or the editor(s). EWA Publishing and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

About volume

Volume title: Proceedings of the 3rd International Conference on Signal Processing and Machine Learning

ISBN：978-1-915371-55-3(Print) / 978-1-915371-56-0(Online)
Editor：Omer Burak Istanbullu
Conference website: http://www.confspml.org
Conference date: 25 February 2023
Series: Applied and Computational Engineering
Volume number: Vol.4
ISSN：2755-2721(Print) / 2755-273X(Online)

© 2024 by the author(s). Licensee EWA Publishing, Oxford, UK. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license. Authors who publish this series agree to the following terms:
1. Authors retain copyright and grant the series right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this series.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the series's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this series.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See Open access policy for details).

References

[1]. Chunwei T Ruibin Z Zhihao W Yong X Wangmeng Z Chen C and Chia-Wen L 2020 Lightweight image super-resolution with enhanced CNN arXiv eess.IV(2020)2007.04344v3

[2]. Chao D, Chen C L, Kaiming H, and Xiaoou T Image super-resolution using deep convolutional networks arXiv cs.CV (2015)1501.00092v3

[3]. Jin Y Shigesumi K and Takio K Fast and accurate image super resolution by deep CNN with skip connection and network in network arXiv cs.CV (2017) 1707.05425v7

[4]. Jiwon K Jung K L and Kyoung M L Deeply-recursive convolutional network for image super-resolution arXiv cs.CV (2016)1501.04491v2

[5]. Christian L Lucas T Ferenc H Jose C Andrew C Alejandro A Andrew A Alykhan T Johannes T Zehan W and Wenzhe S Photo-realistic single image super-resolution using a generative adversarial network arXiv cs.CV (2017)1609.04802v5

[6]. Zhisheng L Juncheng L Hong L Chaoyan H Linlin Z and Tieyong Z Transformer for single image super-resolution arXiv cs.CV (2022)2108.11084v3

[7]. Chao D Chen C L and Xiaoou T Accelerating the super-resolution convolutional neural network arXiv cs.CV (2016)1608.00367v1

[8]. Zhisheng L Juncheng L Hong L Chaoyan H Linlin Z and Tieyong Z Transformer for single image super-resolution arXiv cs.CV (2022) 2108.11084v3

[9]. Jianrui C Hui Z Hongwei Y Zisheng C and Lei Z Toward real-world single image super-resolution: a new benchmark and a new model arXiv cs.CV (2019) 1904.00523

[10]. Jie H Li S Samuel Al Gang S and Enhua W Squeeze-and-excitation networks arXiv cs.CV (2019) 1709.01507v4

[11]. Kaiming H Xiangyu Z Shaoqing R and Jian S Deep residual learning for image recognition arXiv cs.CV (2015)1512.03385v1

[12]. Jiali W Zhengchun L Ian F Won C Rajkumar K and V. Rao K Fast and accurate learned multiresolution dynamical downscaling for precipitation arXiv cs.LG(2021) 2101.06813v1

[13]. Hang Z Orazio G Iuri F and Jan K Loss functions for image restoration with neural networks arXiv cs.LG(2021) 1511.08861v3

[14]. Alain H and Djemel Z Image quality metrics: PSNR vs. SSIM 2010 International Conference on Pattern Recognition

[15]. Honggang C Xiaohai H Linbo Q Yuanyuan W Chao R and Ce Z Real-world single image super-resolution: a brief review arXiv eess.IV(2021 2103.02368v1

[16]. Yang J Wright J Huang T S and Ma Y Image super resolution via sparse representation arXiv IEEE Transactions on Image Processing, vol. 19, no. 11, pp. 2861-2873 (2010)

[17]. Djork-Arne ́ C Thomas U and Sepp H Fast and accurate deep network learning by exponential linear units (ELUS) arXiv cs.LG(2016) 1511.07289v5

[18]. Xuezhe M Apollo: an adaptive parameter-wise diagonal quasi-newton method for nonconvex stochastic optimization arXiv cs.LG (2016)2009.13586v6

[19]. Bevilacqua M Roumy A Guillemot C and Morel M.L.A. Low-complexity single-image super-resolution based on nonnegative neighbor embedding arXiv BMVC. (2012)

[20]. Zeyde R Elad M and Protter M On single image scale-up using sparse-representations arXiv Curves and Surfaces (2012) 711–730

[21]. Martin D Fowlkes C Tal D and Malik J A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics arXiv ICCV Volume 2 (2001) 416–423

For authors
Guide for authors Policies for authors Aims and scope Subscription Open access policy
For editors
Policies for editors Editorial board
For reviewers
Policies for reviewers Privacy policy

Cookies are used by this site.
Copyright © 2024 EWA Publishing, its licensors, and contributors. All rights reserved, including text and data mining, AI training, and similar technologies.