Comparison of Deep Learning Methods for Underwater Image Enhancement

DOI

10.2991/978-94-6463-094-7_45How to use a DOI?

Keywords

Underwater image enhancement; deep learning; fusion; image formation model; comprehensive evaluation

Abstract

Underwater image enhancement is an important process in image processing due to the images often suffering from severe degradation causes by the nature of light and underwater environment. The purpose of this research is to study the existing methods and algorithms for enhancing underwater images. In this paper, we compared 3 different deep learning-based methods (i.e. Water-Net, Shallow-UWnet, Deep Learning and Image Formation Model) for underwater image enhancement. Furthermore, we proposed an enhancement method based on white balance, adaptive gamma correction, sharpening and multi-scale fusion technique. The result of our proposed method is fed into the deep learning-based models. A real-world dataset which is the Underwater Image Enhancement Benchmark (UIEB) dataset is used for the model training and testing. Experimental results show that our proposed method improves the colour hue, image clarity and achieves higher scores in terms PSNR, SSIM and UIQM metrics.

Copyright

© 2022 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)

TY  - CONF
AU  - An’nissa Ariqah Jobli
AU  - Noramiza Hashim
PY  - 2022
DA  - 2022/12/27
TI  - Comparison of Deep Learning Methods for Underwater Image Enhancement
BT  - Proceedings of the International Conference on Computer, Information Technology and Intelligent Computing (CITIC 2022)
PB  - Atlantis Press
SP  - 558
EP  - 571
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-094-7_45
DO  - 10.2991/978-94-6463-094-7_45
ID  - Jobli2022
ER  -