Beyond Single Reference for Training: Underwater Image Enhancement via Comparative Learning

doi:10.1109/TCSVT.2022.3225376

	Beyond Single Reference for Training: Underwater Image Enhancement via Comparative Learning
	Li, Kunqian 1; Wu, Li 1; Qi, Qi 2; Liu, Wenjie 1; Gao, Xiang3 ; Zhou, Liqin 1; Song, Dalei 1,4
发表期刊	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
ISSN	1051-8215
	2023-06-01
卷号	33 期号:6 页码:2561-2576
通讯作者	Song, Dalei(songdalei@ouc.edu.cn)
摘要	Due to the wavelength-dependent light absorption and scattering, the raw underwater images are usually inevitably degraded. Underwater image enhancement (UIE) is of great importance for underwater observation and operation. Data-driven methods, such as deep learning-based UIE approaches, tend to be more applicable to real underwater scenarios. However, the training of deep models is limited by the extreme scarcity of underwater images with enhancement references, resulting in their poor performance in dynamic and diverse underwater scenes. As an alternative, enhancement reference achieved by volunteer voting alleviate the sample shortage to some extent. Since such artificially acquired references are not veritable ground truth, they are far from complete and accurate to provide correct and rich supervision for the enhancement model training. Beyond training with single reference, we propose the first comparative learning framework for UIE problem, namely CLUIE-Net, to learn from multiple candidates of enhancement reference. This new strategy also supports semi-supervised learning mode. Besides, we propose a regional quality-superiority discriminative network (RQSD-Net) as an embedded quality discriminator for the CLUIE-Net. Comprehensive experiments demonstrate the effectiveness of RQSD-Net and the comparative learning strategy for UIE problem. The code, models and new dataset RQSD-UI are available at: https://justwj.github.io/CLUIE-Net.html/.
关键词	Training Image enhancement Visualization Task analysis Generators Deep learning Oceans Underwater image enhancement deep learning convolutional neural network comparative learning blind image quality assessment
DOI	10.1109/TCSVT.2022.3225376
关键词[WOS]	QUALITY ASSESSMENT ; NEURAL-NETWORK ; CHALLENGES ; WATER
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61906177] ; Natural Science Foundation of Shandong Province[ZR2019BF034] ; Fundamental Research Funds for the Central Universities[201964013]
项目资助者	National Natural Science Foundation of China ; Natural Science Foundation of Shandong Province ; Fundamental Research Funds for the Central Universities
WOS研究方向	Engineering
WOS类目	Engineering, Electrical & Electronic
WOS记录号	WOS:001004257300003
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：21[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/53748
专题	中国科学院工业视觉智能装备工程实验室
通讯作者	Song, Dalei
作者单位	1.Ocean Univ China, Coll Engn, Qingdao 266100, Peoples R China 2.Ocean Univ China, Coll Comp Sci & Technol, Qingdao 266100, Peoples R China 3.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 4.Ocean Univ China, Inst Adv Ocean Study, Qingdao 266100, Peoples R China
推荐引用方式 GB/T 7714	Li, Kunqian,Wu, Li,Qi, Qi,et al. Beyond Single Reference for Training: Underwater Image Enhancement via Comparative Learning[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2023,33(6):2561-2576.
APA	Li, Kunqian.,Wu, Li.,Qi, Qi.,Liu, Wenjie.,Gao, Xiang.,...&Song, Dalei.(2023).Beyond Single Reference for Training: Underwater Image Enhancement via Comparative Learning.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,33(6),2561-2576.
MLA	Li, Kunqian,et al."Beyond Single Reference for Training: Underwater Image Enhancement via Comparative Learning".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 33.6(2023):2561-2576.