Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)

<Previous Article In Volume
Next Article In Volume>

Low-Light Image Enhancement based on Zero-DCE and Structural Similarity Loss

Authors
Qiyao Li1, Zhequan Li2, *, Haoyang Wang3
1Detroit Green Technology Institute, Hubei University of Technology, 430068, No.28 Nanli Road, Wuhan, Hubei, China
2College of Biomedical Engineering and Instrument Science, Zhejiang University, 310027, No.38 Zheda Road, Hangzhou, Zhejiang, China
3College of International Education, Shanghai Jian Qiao University, 201306, No.1111 Huchenghuan Road, Shanghai, China
*Corresponding author. Email: lizhequan@zju.edu.cn
Corresponding Author
Zhequan Li
Available Online 27 November 2023.
DOI
10.2991/978-94-6463-300-9_96How to use a DOI?
Keywords
LLIE; UNet3+; image denoising; SSIM loss
Abstract

The computer vision community has become increasingly interested in Low-Light Image Enhancement (LLIE), which tries to transform low-light photos into typically exposed images. The convolutional neural network has advanced quickly, and this has helped the deep learning-based LLIE approaches make a breakthrough in accuracy and visual effects. However, some challenges still remain, especially when dealing with noise from the black color blocks and halo near the boundary of the bright area. In this study, we provide a low-light picture enhancing technique based on the Unet3+ to overcome these problems. Specifically, we first transform DCE-Net in Zero-DCE to Unet3+, which enhances the network's fitting ability. Then, we introduce a denoising module and an SSIM loss, which can improve the qualitative and quantitative metrics of the network. Numerous tests support the effectiveness of our suggested approach, where the normal exposure images produced have a stable brightness and are suitable for a range of scenes.

Copyright
© 2023 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 Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
Series
Advances in Computer Science Research
Publication Date
27 November 2023
ISBN
10.2991/978-94-6463-300-9_96
ISSN
2352-538X
DOI
10.2991/978-94-6463-300-9_96How to use a DOI?
Copyright
© 2023 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  - Qiyao Li
AU  - Zhequan Li
AU  - Haoyang Wang
PY  - 2023
DA  - 2023/11/27
TI  - Low-Light Image Enhancement based on Zero-DCE and Structural Similarity Loss
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 948
EP  - 960
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_96
DO  - 10.2991/978-94-6463-300-9_96
ID  - Li2023
ER  -