Education, Science, Technology, Innovation and Life
Open Access
Sign In

Method for Underwater Image Enhancement Based on Multiscale Feature Fusion

Download as PDF

DOI: 10.23977/jipta.2026.090106 | Downloads: 0 | Views: 7

Author(s)

Ke Hu 1

Affiliation(s)

1 Nanning University, Nanning, Guangxi, 530299, China

Corresponding Author

Ke Hu

ABSTRACT

Light absorption and scattering cause severe color shifts and detail degradation in underwater images, hindering marine exploration and robotic vision applications. To address this issue, this paper presents a design for an underwater image enhancement system based on multiscale feature fusion. The system uses an encoder-decoder approach combined with a pyramid fusion strategy to extract features with different receptive fields through parallel multi-branch convolutions. It also introduces a spatial attention mechanism to perform adaptive weighted fusion and achieve image reconstruction via frequency-domain decomposition. Experiments validate that this method demonstrates excellent performance in color correction, detail preservation, and cross-scene generalization, providing an effective technical solution for underwater vision systems. 

KEYWORDS

Underwater image enhancement; Multiscale feature fusion; Encoder-decoder; Pyramid fusion

CITE THIS PAPER

Ke Hu. Method for Underwater Image Enhancement Based on Multiscale Feature Fusion. Journal of Image Processing Theory and Applications (2026) Vol. 9, No.1, 58-64. DOI: http://dx.doi.org/10.23977/jipta.2026.090106.

REFERENCES

[1] Liu Y, Liu M, Lin S, et al. Underwater image enhancement based on multi-scale feature fusion and attention network[J]. Journal of computer-aided design & computer graphics, 2023, 35(5): 685-695.
[2] Yan X, Liu X, Qu R, et al. A multi-scale feature extraction and attention aggregation network for underwater image enhancement[J]. Expert Systems with Applications, 2025: 129226.
[3] Li M, Liu W, Shao C, et al. Multi-scale feature enhancement method for underwater object detection[J]. Symmetry, 2025, 17(1): 63.
[4] Dai Y, Wang J, Wang H, et al. Underwater image enhancement based on multiscale fusion generative adversarial network[J]. International Journal of Machine Learning and Cybernetics, 2024, 15(4): 1331-1341.
[5] Mi Y, Chi M, Zhang Q, et al. Research on multi-scale fusion image enhancement and improved YOLOv5s lightweight ROV underwater target detection method[J]. Scientific Reports, 2024, 14(1): 28280.
[6] Lu C, Hong L, Fan Y, et al. A Multi-Scale Contextual Fusion Residual Network for Underwater Image Enhancement[J]. Journal of Marine Science and Engineering, 2025, 13(8): 1531.
[7] Sathish K. Deep multiscale fusion and feature enhancement for underwater image dehazing in turbid environments[C]//2025 International Conference on Intelligent Communication Networks and Computational Techniques (ICICNCT). IEEE, 2025: 1-5.
[8] Pan C, Wu Y, Wang X, et al. Dmf-net: Dynamic multiscale feature fusion for underwater image enhancement[J]. Circuits, Systems, and Signal Processing, 2026: 1-28.
[9] Lan R, Guo D, Du P, et al. Multi-scale cross-layer feature interaction GAN for underwater image enhancement[J]. Digital Signal Processing, 2024, 145: 104354.
[10] Cao W, Zhang W. Multi scale perceptual underwater image enhancement based on feature fusion network[C]//2024 IEEE 6th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC). IEEE, 2024, 6: 1128-1136.

Downloads: 3285
Visits: 291055

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.