# Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding

1 Nanyang Technological University, Singapore
2 CSIRO, Australia
3 City University of Hong Kong, Hong Kong
4 Beijing Jiaotong University, China
5 Nankai University, Tianjin, China
6 Institute of Information Engineering, Chinese Academy of Sciences, China

## Abstract

Underwater images suffer from color casts and low contrast due to wavelength- and distance-dependent attenuation and scattering. To solve these two degradation issues, we present an underwater image enhancement network via medium transmission-guided multi-color space embedding, called Ucolor. Concretely, we first propose a multi-color space encoder network, which enriches the diversity of feature representations by incorporating the characteristics of different color spaces into a unified structure. Coupled with an attention mechanism, the most discriminative features extracted from multiple color spaces are adaptively integrated and highlighted. Inspired by underwater imaging physical models, we design a medium transmission (indicating the percentage of the scene radiance reaching the camera)-guided decoder network to enhance the response of network towards quality-degraded regions. As a result, our network can effectively improve the visual quality of underwater images by exploiting multiple color spaces embedding and the advantages of both physical model-based and learning-based methods. Extensive experiments demonstrate that our Ucolor achieves superior performance against state-of-the-art methods in terms of both visual quality and quantitative metrics.

## Highlights

1. We propose a multi-color space encoder network coupled with an attention mechanism for incorporating the characteristics of different color spaces into a unified structure and adaptively selecting the most representative features.

2. We propose a medium transmission-guided decoder network to enforce the network to pay more attention to quality-degraded regions. It explores the complementary merits between domain knowledge of underwater imaging and deep neural networks.

3. Our Ucolor achieves state-of-the-art performance on several recent benchmarks in terms of both visual quality and quantitative metrics.

## Method

Overview of the architecture of Ucolor. Our Ucolor consists of a multi-color space encoder network and a medium transmission-guided decoder network. In our method, we normalize the values of the medium transmission map to [0,1] and feed the reverse medium transmission map (denoted as RMT) to the medium transmission guidance module.

## Materials

 Paper Supplementary Code

## Citation

@Article{Ucolor,
author ={Li, Chongyi and Anwar, Saeed and Hou, Junhui and Cong, Runmin and Guo, Chunle and Ren, Wenqi},
title = {Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding},
journal = {IEEE Transactions on Image Processing},
Volume={30},
pape={4985-5000},
year = {2021},
doi={10.1109/TIP.2021.3076367}
}