Underwater images often suffer from severe visual degradation due to the complex imaging and transmission conditions in underwater environments. In resource-constrained underwater systems, image enhancement and transmission must jointly consider restoration quality, communication efficiency, and energy consumption. In particular, autonomous underwater vehicle (AUV) swarms require lightweight solutions that balance computational complexity with limited bandwidth and energy resources. To address these challenges, we propose a lightweight joint image enhancement and transmission framework tailored for AUV swarms. The proposed framework integrates a target-oriented multi-space feature encoding network to extract robust representations from degraded images, followed by a deep reinforcement learning (DRL)-based bandwidth allocation module that optimizes feature transmission under constrained communication resources. Furthermore, a DRL-driven link selection strategy is designed to adaptively determine energy-efficient communication links, thereby improving overall system energy efficiency. Extensive simulations demonstrate that the proposed framework significantly improves image restoration speed and quality under diverse channel conditions. Moreover, as the swarm size increases, the proposed method achieves more balanced energy consumption among AUVs, leading to enhanced system-wide energy efficiency and prolonged operational lifetime.
Zhang et al. (Sat,) studied this question.