Cycle-Retinex: Unpaired Low-Light Image Enhancement via Retinex-Inline CycleGAN

被引:18
作者
Wu, Kangle [1 ]
Huang, Jun [1 ]
Ma, Yong [1 ]
Fan, Fan [1 ]
Ma, Jiayi [1 ]
机构
[1] Wuhan Univ, Elect Informat Sch, Wuhan 430072, Peoples R China
基金
中国国家自然科学基金;
关键词
Low light enhancement; CycleGAN; Retinex theory; unsupervised learning; HISTOGRAM EQUALIZATION; FUSION;
D O I
10.1109/TMM.2023.3278385
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Low-light image enhancement aims to recover normal-light images from the images captured under dim environments. Most existing methods could just improve the light appearance globally whereas failing to handle other degradation such as dense noise, color offset and extremely low-light. Moreover, unsupervised methods proposed in recent years lack reliable physical model as the basis, thus universality is greatly limited. To address these problems, we propose a novel low-light image enhancement method via Retinex-inline cycle-consistent generative adversarial network named Cycle-Retinex, whose training is totally dependent on unpaired datasets. Specifically, we organically combine Retinex theory with CycleGAN, by which we decouple low-light image enhancement task into two sub-tasks, i.e. illumination map enhancement and reflectance map restoration. Retinex theory helps CycleGAN simplify low-light image enhancement problem and CycleGAN provides synthetic paired images to guide the training of Retinex decomposition network. We further introduce a self-augmented method to address the color distortion and noise problem, thus making the network learn to enhance low-light images adaptively. Extensive experiments show that the proposed method can achieve promising results.
引用
收藏
页码:1213 / 1228
页数:16
相关论文
共 47 条
[1]  
[Anonymous], 2018, PROC EUR C COMPUT VI, DOI [DOI 10.1007/978-3-030-01234-2_49, 10.1007/978-3-030-01234-2_49]
[2]   Learning a Deep Single Image Contrast Enhancer from Multi-Exposure Images [J].
Cai, Jianrui ;
Gu, Shuhang ;
Zhang, Lei .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2018, 27 (04) :2049-2062
[3]   Minimum mean brightness error bi-histogram equalization in contrast enhancement [J].
Chen, SD ;
Ramli, R .
IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, 2003, 49 (04) :1310-1319
[4]   Deep Photo Enhancer: Unpaired Learning for Image Enhancement from Photographs with GANs [J].
Chen, Yu-Sheng ;
Wang, Yu-Ching ;
Kao, Man-Hsin ;
Chuang, Yung-Yu .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :6306-6314
[5]   StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation [J].
Choi, Yunjey ;
Choi, Minje ;
Kim, Munyoung ;
Ha, Jung-Woo ;
Kim, Sunghun ;
Choo, Jaegul .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :8789-8797
[6]   A Probabilistic Method for Image Enhancement With Simultaneous Illumination and Reflectance Estimation [J].
Fu, Xueyang ;
Liao, Yinghao ;
Zeng, Delu ;
Huang, Yue ;
Zhang, Xiao-Ping ;
Ding, Xinghao .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2015, 24 (12) :4965-4977
[7]   Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement [J].
Guo, Chunle ;
Li, Chongyi ;
Guo, Jichang ;
Loy, Chen Change ;
Hou, Junhui ;
Kwong, Sam ;
Cong, Runmin .
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, :1777-1786
[8]   LIME: Low-Light Image Enhancement via Illumination Map Estimation [J].
Guo, Xiaojie ;
Li, Yu ;
Ling, Haibin .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2017, 26 (02) :982-993
[9]   Towards Low Light Enhancement With RAW Images [J].
Huang, Haofeng ;
Yang, Wenhan ;
Hu, Yueyu ;
Liu, Jiaying ;
Duan, Ling-Yu .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 31 :1391-1405
[10]   Efficient Contrast Enhancement Using Adaptive Gamma Correction With Weighting Distribution [J].
Huang, Shih-Chia ;
Cheng, Fan-Chieh ;
Chiu, Yi-Sheng .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2013, 22 (03) :1032-1041