Cargando…

Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks

A network structure (DRSN-GAN) is proposed for image motion deblurring that combines a deep residual shrinkage network (DRSN) with a generative adversarial network (GAN) to address the issues of poor noise immunity and low generalizability in deblurring algorithms based solely on GANs. First, an end...

Descripción completa

Detalles Bibliográficos
Autores principales: Jiang, Wenbo, Liu, Anshun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Hindawi 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8799329/
https://www.ncbi.nlm.nih.gov/pubmed/35096042
http://dx.doi.org/10.1155/2022/5605846
_version_ 1784642045725376512
author Jiang, Wenbo
Liu, Anshun
author_facet Jiang, Wenbo
Liu, Anshun
author_sort Jiang, Wenbo
collection PubMed
description A network structure (DRSN-GAN) is proposed for image motion deblurring that combines a deep residual shrinkage network (DRSN) with a generative adversarial network (GAN) to address the issues of poor noise immunity and low generalizability in deblurring algorithms based solely on GANs. First, an end-to-end approach is used to recover a clear image from a blurred image, without the need to estimate a blurring kernel. Next, a DRSN is used as the generator in a GAN to remove noise from the input image while learning residuals to improve robustness. The BN and ReLU layers in the DRSN were moved to the front of the convolution layer, making the network easier to train. Finally, deblurring performance was verified using the GoPro, Köhler, and Lai datasets. Experimental results showed that deblurred images were produced with more subjective visual effects and a higher objective evaluation, compared with algorithms such as MPRNet. Furthermore, image edge and texture restoration effects were improved along with image quality. Our model produced slightly higher PSNR and SSIM values than the latest MPRNet, as well as increased YOLO detection accuracy. The number of required parameters in the DRSN-GAN was also reduced by 21.89%.
format Online
Article
Text
id pubmed-8799329
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Hindawi
record_format MEDLINE/PubMed
spelling pubmed-87993292022-01-29 Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks Jiang, Wenbo Liu, Anshun Comput Intell Neurosci Research Article A network structure (DRSN-GAN) is proposed for image motion deblurring that combines a deep residual shrinkage network (DRSN) with a generative adversarial network (GAN) to address the issues of poor noise immunity and low generalizability in deblurring algorithms based solely on GANs. First, an end-to-end approach is used to recover a clear image from a blurred image, without the need to estimate a blurring kernel. Next, a DRSN is used as the generator in a GAN to remove noise from the input image while learning residuals to improve robustness. The BN and ReLU layers in the DRSN were moved to the front of the convolution layer, making the network easier to train. Finally, deblurring performance was verified using the GoPro, Köhler, and Lai datasets. Experimental results showed that deblurred images were produced with more subjective visual effects and a higher objective evaluation, compared with algorithms such as MPRNet. Furthermore, image edge and texture restoration effects were improved along with image quality. Our model produced slightly higher PSNR and SSIM values than the latest MPRNet, as well as increased YOLO detection accuracy. The number of required parameters in the DRSN-GAN was also reduced by 21.89%. Hindawi 2022-01-21 /pmc/articles/PMC8799329/ /pubmed/35096042 http://dx.doi.org/10.1155/2022/5605846 Text en Copyright © 2022 Wenbo Jiang and Anshun Liu. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Research Article
Jiang, Wenbo
Liu, Anshun
Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks
title Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks
title_full Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks
title_fullStr Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks
title_full_unstemmed Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks
title_short Image Motion Deblurring Based on Deep Residual Shrinkage and Generative Adversarial Networks
title_sort image motion deblurring based on deep residual shrinkage and generative adversarial networks
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8799329/
https://www.ncbi.nlm.nih.gov/pubmed/35096042
http://dx.doi.org/10.1155/2022/5605846
work_keys_str_mv AT jiangwenbo imagemotiondeblurringbasedondeepresidualshrinkageandgenerativeadversarialnetworks
AT liuanshun imagemotiondeblurringbasedondeepresidualshrinkageandgenerativeadversarialnetworks