Proceedings of the 2023 International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2023)

<Previous Article In Volume
Next Article In Volume>

Current Study on Image Restoration Leveraging CNNs and GANs

Authors
Feng Cai1, Jingxu Peng2, *, Peng Zhou3
1South China University of Technology, Panyu District, Guangzhou, 510006, China
2Tianjin University of Technology, Xiqing District, Tianjin, 300384, China
3Chengdu College of University of Electronic Science and Technology of China, Western High-Tech District, Chengdu, 611731, China
*Corresponding author. Email: lukasy@stud.tjut.edu.cn
Corresponding Author
Jingxu Peng
Available Online 14 February 2024.
DOI
10.2991/978-94-6463-370-2_35How to use a DOI?
Keywords
CNNs; GANs; Image Restoration; DnCNN; SRGAN
Abstract

In recent years, advances in image restoration have prominently featured Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs). Specifically, in seismic image denoising, the DnCNN model, rooted in CNN, employs residual learning combined with batch normalization for a streamlined approach to denoising. Alternatively, SRGAN, leveraging GAN and super-resolution techniques, trains its generative model using a perceptual loss function, accentuating perceptual disparities between generated and authentic images to boost visual quality. Recognizing the importance of preserving salient seismic data, the Constrained-DnCNN model has been introduced, refining seismic data interpretation. Additionally, the CR-SRGAN model, targeting super-resolution of artifacts and color restoration, deviates from conventional training datasets acquired via high-resolution image interpolation and downsampling. Upon comparative analysis, the DCGAN, amalgamating features of CNN and GAN, stands out for harnessing CNN’s robust feature extraction. This bolsters the model’s capacity to adeptly fit real data distributions. DCGAN’s potential spans aspects like multimodal fusion, model interpretability, trustworthiness, and image restoration quality, marking it a promising research avenue.

Copyright
© 2024 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the 2023 International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2023)
Series
Advances in Intelligent Systems Research
Publication Date
14 February 2024
ISBN
978-94-6463-370-2
ISSN
1951-6851
DOI
10.2991/978-94-6463-370-2_35How to use a DOI?
Copyright
© 2024 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - Feng Cai
AU  - Jingxu Peng
AU  - Peng Zhou
PY  - 2024
DA  - 2024/02/14
TI  - Current Study on Image Restoration Leveraging CNNs and GANs
BT  - Proceedings of the 2023 International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2023)
PB  - Atlantis Press
SP  - 321
EP  - 333
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-370-2_35
DO  - 10.2991/978-94-6463-370-2_35
ID  - Cai2024
ER  -