CARNet: Collaborative Adversarial Resilience for Robust Underwater Image Enhancement and Perception
IEEE Transactions on Circuits and Systems for Video Technology · 2026-11 · 36 (11) · 16832-16845
DOI: 10.1109/tcsvt.2026.3715053 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-11. Verfasst von Zengxi Zhang und 3 weiteren Autor:innen. Die Arbeit nennt 73 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Liaoning Provincial Natural Science Foundation.
Abstract (Originalsprache)
Due to the uneven absorption of different light wavelengths in aquatic environments, underwater images suffer from low visibility and clear color deviations. With the advancement of autonomous underwater vehicles, extensive research has been conducted on learning-based underwater enhancement algorithms. These works can generate visually pleasing enhanced images and mitigate the adverse effects of degraded images on subsequent perception tasks. However, learning-based methods are susceptible to the inherent fragility of adversarial attacks, causing significant disruption in enhanced results. In this work, we make an early attempt to explore adversarial robustness in underwater image processing and introduce a collaborative adversarial resilience network, dubbed CARNet, for underwater image enhancement and subsequent detection tasks. Concretely, we first introduce an invertible network with strong perturbation-perceptual abilities to isolate attacks from underwater images, preventing interference with visual quality enhancement and perceptual tasks. Furthermore, an attack pattern discriminator is introduced to adaptively identify and eliminate various types of attacks. Additionally, we propose a bilevel attack optimization strategy to heighten the robustness of the network against different types of attacks under the collaborative adversarial training of vision-driven and perception-driven attacks. Extensive experiments demonstrate that the proposed method produces visually pleasing enhanced images and achieves robust detection results under both clean and adversarially attacked settings.
Förderung
- National Natural Science Foundation of China
- Liaoning Provincial Natural Science Foundation
Zitieren
APA
Zhang, Z., Shi, Z., Jiang, Z., & Liu, J. (2026). CARNet: Collaborative Adversarial Resilience for Robust Underwater Image Enhancement and Perception. IEEE Transactions on Circuits and Systems for Video Technology, 36(11), 16832-16845. https://doi.org/10.1109/tcsvt.2026.3715053
Vancouver
Zhang Z, Shi Z, Jiang Z, Liu J. CARNet: Collaborative Adversarial Resilience for Robust Underwater Image Enhancement and Perception. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(11):16832-16845. doi:10.1109/tcsvt.2026.3715053
BibTeX
@article{Zhang2026,
author = {Zhang, Zengxi and Shi, Zeru and Jiang, Zhiying and Liu, Jinyuan},
title = {CARNet: Collaborative Adversarial Resilience for Robust Underwater Image Enhancement and Perception},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2026},
volume = {36},
number = {11},
pages = {16832--16845},
doi = {10.1109/tcsvt.2026.3715053}
}
EUVIMED