🤖 AI Summary
Underwater images suffer from low visibility and severe color distortion due to spectral-selective light absorption; while learning-based enhancement methods improve visual quality, they are vulnerable to adversarial attacks, degrading downstream detection performance. To address this, we propose a collaborative adversarial training framework integrating a reversible perturbation-aware network and an attack-pattern discriminator. We introduce the first vision-driven and perception-driven dual-level adversarial optimization strategy, enabling adaptive robust suppression against diverse attack types. The enhancement process is modeled using an invertible neural network, jointly ensuring fidelity and reversibility. Under maintained visual quality of enhanced images, our method achieves an average 6.71% mAP improvement over state-of-the-art underwater object detection methods, significantly enhancing the robustness of underwater perception systems in adversarial environments.
📝 Abstract
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 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 outputs visually appealing enhancement images and performs an average 6.71% higher detection mAP than state-of-the-art methods.
