Enhancement of Underwater Images with Retinex Transmission Map and Adaptive Color Correction

Underwater images often suffer from low contrast, visibility, and color deviation. In this work, we propose a hybrid underwater enhancement method consisting of addressing an inverse problem with novel Retinex transmission map estimation adaptive correction. does not rely on channel priors aims to decouple the unknown background light, thus avoiding error accumulation problem. To end, global white balance is performed before estimating using multi-scale Retinex. further improve performance, design correction which cleverly chooses between two procedures prevents stretching imbalance. Quantitative qualitative comparisons our state-of-the-art image methods demonstrate superiority proposed method. It achieves best performance in terms full-reference quality assessment. addition, it also superior non-reference evaluation.