An underwater image enhancement model for domain adaptation

Underwater imaging has been suffering from color imbalance, low contrast, and low-light environment due to strong spectral attenuation of light in the water. Owing its complex physical mechanism, enhancing underwater quality based on deep learning method well-developed recently. However, individual studies use different image datasets, leading generalization ability other water conditions. To solve this domain adaptation problem, paper proposes an enhancement scheme that combines individually degraded images publicly available datasets for adaptation. Firstly, dataset fitting model (UDFM) is proposed merge localized into a combined one. Then (UIEM) developed base open clear pairs dataset. The experiment proves can be recovered by only collecting at some specific sea area. Thus, study, problem could solved with increase collected various areas. Also, supposed become more robust. code https://github.com/fanren5599/UIEM .