Deep Color-Corrected Multiscale Retinex Network for Underwater Image Enhancement

颜色恒定性 人工智能 计算机科学 水下 卷积神经网络 计算机视觉 深度学习 颜色校正 块(置换群论) 能见度 稳健性(进化) 灰度 图像(数学) 数学 生物化学 海洋学 物理 几何学 化学 光学 基因 地质学
作者
Hao Qi,Huiyu Zhou,Junyu Dong,Xinghui Dong
出处
期刊:IEEE Transactions on Geoscience and Remote Sensing [Institute of Electrical and Electronics Engineers]
卷期号:62: 1-13 被引量:7
标识
DOI:10.1109/tgrs.2023.3338611
摘要

The acquisition of high-quality underwater images is of great importance to ocean exploration activities. However, images captured in the underwater environment often suffer from degradation due to complex imaging conditions, leading to various issues, such as color cast, low contrast and low visibility. Although many traditional methods have been used to address these issues, they usually lack robustness in diverse underwater scenes. On the other hand, deep learning techniques struggle to generalize to unseen images, due to the challenge of learning the complicated degradation process. Inspired by the success achieved by the Retinex-based methods, we decompose the Underwater Image Enhancement (UIE) task into two consecutive procedures, including color correction and visibility enhancement, and introduce a novel deep Color-Corrected Multi-scale Retinex Network (CCMSR-Net). With regard to the two procedures, this network comprises a Color Correction subnetwork (CC-Net) and a Multi-scale Retinex subnetwork (MSR-Net), which are built on top of the Hybrid Convolution-Axial Attention Block (HCAAB) that we design. Thanks to this block, the CCMSR-Net is able to efficiently capture local characteristics and the global context. Experimental results show that the CCMSR-Net outperforms, or at least performs comparably to, 11 baselines across five test sets. We believe that these promising results are due to the effective combination of color correction methods and the multi-scale Retinex model, achieved by jointly exploiting Convolutional Neural Networks (CNNs) and Transformers.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
SciGPT应助科研通管家采纳,获得10
刚刚
慕青应助故意的觅珍采纳,获得10
刚刚
今后应助科研通管家采纳,获得10
刚刚
锅包肉完成签到 ,获得积分10
刚刚
我是老大应助科研通管家采纳,获得10
刚刚
科研通AI2S应助科研通管家采纳,获得10
刚刚
刚刚
wanci应助科研通管家采纳,获得10
1秒前
丘比特应助科研通管家采纳,获得10
1秒前
SYLH应助科研通管家采纳,获得10
1秒前
1秒前
drfwjuikesv应助科研通管家采纳,获得10
1秒前
S先生完成签到,获得积分10
1秒前
1秒前
Hello应助科研通管家采纳,获得10
1秒前
思源应助科研通管家采纳,获得10
1秒前
仔拉发布了新的文献求助10
1秒前
JamesPei应助科研通管家采纳,获得10
1秒前
脑洞疼应助科研通管家采纳,获得10
1秒前
SYLH应助科研通管家采纳,获得10
1秒前
Jasper应助科研通管家采纳,获得10
1秒前
bkagyin应助科研通管家采纳,获得10
1秒前
ED应助科研通管家采纳,获得10
1秒前
深情安青应助科研通管家采纳,获得10
1秒前
SYLH应助科研通管家采纳,获得10
2秒前
2秒前
学海行舟完成签到 ,获得积分10
2秒前
蛋蛋姐姐完成签到,获得积分10
2秒前
li给li的求助进行了留言
2秒前
something完成签到,获得积分10
2秒前
Fayth完成签到,获得积分10
3秒前
zimu012发布了新的文献求助10
3秒前
shelly发布了新的文献求助10
4秒前
热心乌完成签到,获得积分0
4秒前
健忘的小懒虫完成签到,获得积分10
4秒前
可爱的汤圆完成签到,获得积分10
5秒前
6秒前
山猫完成签到,获得积分10
6秒前
萌123完成签到,获得积分10
6秒前
li发布了新的文献求助10
6秒前
高分求助中
Picture Books with Same-sex Parented Families: Unintentional Censorship 1500
A new approach to the extrapolation of accelerated life test data 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
基于可调谐半导体激光吸收光谱技术泄漏气体检测系统的研究 310
Composite Predicates in English 300
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3981923
求助须知:如何正确求助?哪些是违规求助? 3525640
关于积分的说明 11227738
捐赠科研通 3263494
什么是DOI,文献DOI怎么找? 1801502
邀请新用户注册赠送积分活动 879889
科研通“疑难数据库(出版商)”最低求助积分说明 807608