Low-light image enhancement network guided by reflection prior map

doi:10.11772/j.issn.1001-9081.2025020222

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (2): 546-554.DOI: 10.11772/j.issn.1001-9081.2025020222

• Multimedia computing and computer simulation • Previous Articles

Low-light image enhancement network guided by reflection prior map

Rifeng ZHANG, Guangming LI(), Yurong OUYANG

School of Computer Science and Technology，Dongguan University of Technology，Dongguan Guangdong 523808，China

Received:2025-03-06 Revised:2025-06-15 Accepted:2025-06-23 Online:2025-08-08 Published:2026-02-10
Contact: Guangming LI
About author:ZHANG Rifeng， born in 2000， M. S. candidate. His research interests include computer vision， low-light image enhancement.
LI Guangming， born in 1968， M. S.， distinguished professor. His research interests include embedded systems， artificial intelligence. Email:ligm@dgut.edu.cn
OUYANG Yurong， born in 1995， M. S. candidate. His research interests include computer vision， image dehazing.
Supported by:
Guangdong Natural Science Foundation(2023A1515011307)

反射先验图引导的低光图像增强网络

张日丰, 李广明(), 欧阳裕荣

东莞理工学院计算机科学与技术学院，广东东莞 523808

通讯作者: 李广明
作者简介:张日丰（2000—），男，广东茂名人，硕士研究生，CCF会员，主要研究方向：计算机视觉、低光图像增强
李广明（1968—），男，河南信阳人，特聘教授，硕士，CCF会员，主要研究方向：嵌入式系统、人工智能 Email:ligm@dgut.edu.cn
欧阳裕荣（1995—），男，广东佛山人，硕士研究生，主要研究方向：计算机视觉、图像去雾。
基金资助:
广东省自然科学基金资助项目(2023A1515011307)

Abstract

Abstract:

In recent years， the low-light image enhancement methods based on deep learning are inspired by Retinex theory. First， the illumination map is estimated to adjust the brightness， and then the reflectance is restored to achieve low-light enhancement. Therefore， by analyzing the similarity between the low-light scene reflection map and reference reflection map， a low-light image enhancement Network guided by Reflection Prior map （RP-Net） was proposed. Firstly， the similar reflection map was generated by decomposing in Lab color space， and an Reflection Prior feature Adaptive Extractor （RPAE） was designed to re-encode and filter the guiding features from the similar reflection map in the backbone network at different scales. Then， the guiding information was injected into the backbone network through the designed Reflection Prior feature-Guided attention Block （RPGB）. In addition， aiming at the limitations of traditional pixel-by-pixel L1 loss， a harmonic loss function of frequency domain was designed from the perspective of frequency domain analysis， so as to optimize the enhancement effect from the global spectral distribution. Experimental results on LOLv1， LOLv2 and LSRW datasets show that the proposed method is superior to the existing mainstream methods in Structural Similarity （SSIM）， and has the Peak Signal-to-Noise Ratio （PSNR） 1.29 dB and 2.08 dB higher than that of Retinexformer and SAFNet （Spatial And Frequency Network）， on the LOLv2-syn and LSRW datasets respectively， and performs well in balancing color fidelity and enhancement effect.

Key words: low-light image enhancement, Lab color space, Retinex theory, attention mechanism, frequency domain analysis

摘要：

近年来，基于深度学习的低光图像增强方法受Retinex理论的启发，先估计照明图调整亮度，再恢复反射率以实现低光增强。因此，通过分析低光场景反射图与参考反射图的相似度，提出一种反射先验图引导的低光图像增强网络（RP-Net）。首先，在Lab色彩空间分解出相似反射图，并设计反射先验特征自适应提取器（RPAE）在主干网络以不同尺度从相似反射图中重新编码和筛选引导特征；其次，通过设计的反射先验特征引导注意力块（RPGB）将引导信息注入主干网络。此外，针对传统逐像素L1损失的局限性，从频域分析的视角出发，设计一种频域调和损失函数，以从全局频谱分布优化增强效果。在LOLv1、LOLv2和LSRW数据集上的实验结果表明，所提方法在结构相似性（SSIM）上优于现有主流方法，在LOLv2-syn和LSRW数据集上得到的峰值信噪比（PSNR）相较于Retinexformer和SAFNet （Spatial And Frequency Network）分别提高了1.29 dB和2.08 dB，并且在色彩保真和增强效果的平衡上表现出色。

关键词: 低光图像增强, Lab色彩空间, Retinex理论, 注意力机制, 频域分析

CLC Number:

TP391

Rifeng ZHANG, Guangming LI, Yurong OUYANG. Low-light image enhancement network guided by reflection prior map[J]. Journal of Computer Applications, 2026, 46(2): 546-554.

张日丰, 李广明, 欧阳裕荣. 反射先验图引导的低光图像增强网络[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 546-554.

Figures/Tables 14

References 30

[1]	彭大鑫，甄彤，李智慧. 低光照图像增强研究方法综述［J］. 计算机工程与应用， 2023， 59（18）： 14-27.
	PENG D X， ZHEN T， LI Z H. Survey of research methods for low light image enhancement［J］. Computer Engineering and Applications， 2023， 59（18）： 14-27.
[2]	ABDULLAH-AL-WADUD M， KABIR M H， DEWAN M A A， et al. A dynamic histogram equalization for image contrast enhancement［J］. IEEE Transactions on Consumer Electronics， 2007， 53（2）： 593-600.
[3]	RAO R B. Dynamic histogram equalization for contrast enhancement for digital images［J］. Applied Soft Computing， 2020， 89： No.106114.
[4]	DALE-JONES R， TJAHJADI T. A study and modification of the local histogram equalization algorithm［J］. Pattern Recognition， 1993， 26（9）： 1373-1381.
[5]	HUANG S C， CHENG F C， CHIU Y S. Efficient contrast enhancement using adaptive gamma correction with weighting distribution［J］. IEEE Transactions on Image Processing， 2013， 22（3）： 1032-1041.
[6]	LI X， LIU M， LING Q. Pixel-wise gamma correction mapping for low-light image enhancement［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2024， 34（2）： 681-694.
[7]	LAND E H. The Retinex theory of color vision［J］. Scientific American， 1977， 237（6）： 108-128.
[8]	LORE K G， AKINTAYO A， SARKAR S. LLNet： a deep autoencoder approach to natural low-light image enhancement［J］. Pattern Recognition， 2017， 61： 650-662.
[9]	WANG R， ZHANG Q， FU C W， et al. Underexposed photo enhancement using deep illumination estimation［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 6842-6850.
[10]	ZAMIR S W， ARORA A， KHAN S， et al. Learning enriched features for real image restoration and enhancement［C］// Proceedings of the 2020 European Conference on Computer Vision， LNCS 12355. Cham： Springer， 2020： 492-511.
[11]	田会娟，蔡敏鹏，关涛，等. 基于YCbCr颜色空间的Retinex低照度图像增强方法研究［J］. 光子学报， 2020， 49（2）： No.0210002.
	TIAN H J， CAI M P， GUAN T， et al. Low-light image enhancement method using Retinex method based on YCbCr color space［J］. Acta Photonica Sinica， 2020， 49（2）： No.0210002.
[12]	LIU R， MA L， ZHANG J， et al. Retinex-inspired unrolling with cooperative prior architecture search for low-light image enhancement［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2021： 10556-10565.
[13]	WEI C， WANG W， YANG W， et al. Deep Retinex decomposition for low-light enhancement［C］// Proceedings of the 2018 British Machine Vision Conference. Durham： BMVA Press， 2018： No.451.
[14]	ZHANG Y， ZHANG J， GUO X. Kindling the darkness： a practical low-light image enhancer［C］// Proceedings of the 27th ACM International Conference on Multimedia. New York： ACM， 2019： 1632-1640.
[15]	ZHANG Y， GUO X， MA J， et al. Beyond brightening low-light images［J］. International Journal of Computer Vision， 2021， 129（4）： 1013-1037.
[16]	CAI Y， BIAN H， LIN J， et al. Retinexformer： one-stage Retinex-based Transformer for low-light image enhancement［C］// Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2023： 12470-12479.
[17]	WANG Y， WAN R， YANG W， et al. Low-light image enhancement with normalizing flow［C］// Proceedings of the 36th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2022： 2604-2612.
[18]	李大海，王忠华，王振东. 结合空间域和频域信息的双分支低光照图像增强网络［J］. 计算机应用， 2024， 44（7）： 2175-2182.
	LI D H， WANG Z H， WANG Z D. Dual-branch low-light image enhancement network combining spatial and frequency domain information［J］. Journal of Computer Applications， 2024， 44（7）： 2175-2182.
[19]	GUO C， LI C， GUO J， et al. Zero-reference deep curve estimation for low-light image enhancement［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 1777-1786.
[20]	JIANG Y， GONG X， LIU D， et al. EnlightenGAN： deep light enhancement without paired supervision［J］. IEEE Transactions on Image Processing， 2021， 30： 2340-2349.
[21]	YANG W， WANG W， HUANG H， et al. Sparse gradient regularized deep retinex network for robust low-light image enhancement［J］. IEEE Transactions on Image Processing， 2021， 30： 2072-2086.
[22]	WANG Z， BOVIK A C， SHEIKH H R， et al. Image quality assessment： from error visibility to structural similarity［J］. IEEE Transactions on Image Processing， 2004， 13（4）： 600-612.
[23]	YAN Q， FENG Y， ZHANG C， et al. HVI： a new color space for low-light image enhancement［C］// Proceedings of the 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2025： 5678-5687.
[24]	FUOLI D， VAN GOOL L， TIMOFTE R. Fourier space losses for efficient perceptual image super-resolution［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 2340-2349.
[25]	CUI Y， REN W， CAO X， et al. Revitalizing convolutional network for image restoration［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2024， 46（12）： 9423-9438.
[26]	KINGMA D P， BA J L. Adam： a method for stochastic optimization［EB/OL］. ［2025-06-14］..
[27]	LOSHCHILOV I， HUTTER F. SGDR： Stochastic gradient descent with warm restarts［EB/OL］. ［2025-06-14］..
[28]	ZAMIR S W， ARORA A， KHAN S， et al. Restormer： efficient Transformer for high-resolution image restoration［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2022： 5718-5729.
[29]	XU X， WANG R， FU C W， et al. SNR-aware low-light image enhancement［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2022： 17693-17703.
[30]	WANG T， ZHANG K， SHEN T， et al. Ultra-high-definition low-light image enhancement： a benchmark and Transformer-based method［C］// Proceedings of the 37th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2023： 2654-2662.

方法	复杂度		LOLv1		LOLv2-real		LOLv2-syn
方法	GFLOPs	Params/10⁶	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
RetinexNet^［13］	0.840	584.470	16.77	0.560	16.09	0.401	17.13	0.762
Zero-DCE^［19］	0.075	4.830	14.86	0.559	16.06	0.580	17.71	0.810
EnlightenGAN^［20］	61.010	114.350	17.48	0.651	18.23	0.617	16.57	0.734
RUAS^［12］	0.830	0.003	18.23	0.720	18.37	0.723	16.55	0.652
KinD^［14］	34.990	8.020	20.86	0.790	14.74	0.641	13.29	0.578
Restormer^［28］	144.250	26.130	22.43	0.823	19.94	0.827	21.41	0.830
MIRNet^［10］	785.000	31.760	24.14	0.830	20.02	0.820	21.94	0.876
SNR-Net^［29］	26.350	4.010	24.61	0.842	21.48	0.849	24.14	0.928
Retinexformer^［16］	15.570	1.610	25.16	0.845	22.79	0.840	25.67	0.930
HVI-CIDNet^［23］	8.130	1.970	23.23	0.840	20.37	0.846	25.13	0.939
RP-Net	22.060	2.060	24.37	0.857	22.90	0.868	26.96	0.944

方法	复杂度		LOLv1		LOLv2-real		LOLv2-syn
方法	GFLOPs	Params/10⁶	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
RetinexNet^［13］	0.840	584.470	16.77	0.560	16.09	0.401	17.13	0.762
Zero-DCE^［19］	0.075	4.830	14.86	0.559	16.06	0.580	17.71	0.810
EnlightenGAN^［20］	61.010	114.350	17.48	0.651	18.23	0.617	16.57	0.734
RUAS^［12］	0.830	0.003	18.23	0.720	18.37	0.723	16.55	0.652
KinD^［14］	34.990	8.020	20.86	0.790	14.74	0.641	13.29	0.578
Restormer^［28］	144.250	26.130	22.43	0.823	19.94	0.827	21.41	0.830
MIRNet^［10］	785.000	31.760	24.14	0.830	20.02	0.820	21.94	0.876
SNR-Net^［29］	26.350	4.010	24.61	0.842	21.48	0.849	24.14	0.928
Retinexformer^［16］	15.570	1.610	25.16	0.845	22.79	0.840	25.67	0.930
HVI-CIDNet^［23］	8.130	1.970	23.23	0.840	20.37	0.846	25.13	0.939
RP-Net	22.060	2.060	24.37	0.857	22.90	0.868	26.96	0.944

方法	PSNR/dB	SSIM	方法	PSNR/dB	SSIM
RetinexNet^［13］	15.90	0.373	LLFormer^［30］	17.16	0.522
Zero-DCE^［19］	15.83	0.466	SAFNet^［18］	17.23	0.550
EnlightenGAN^［20］	16.31	0.469	RP-Net	19.31	0.580
KinD++^［15］	16.47	0.492

方法	PSNR/dB	SSIM	方法	PSNR/dB	SSIM
RetinexNet^［13］	15.90	0.373	LLFormer^［30］	17.16	0.522
Zero-DCE^［19］	15.83	0.466	SAFNet^［18］	17.23	0.550
EnlightenGAN^［20］	16.31	0.469	RP-Net	19.31	0.580
KinD++^［15］	16.47	0.492

Loss	PSNR	Loss	PSNR
L_spa	24.11	L_spa+0.2×L_fre	23.92
L_spa+0.05×L_fre	24.29	L_spa+0.1×L_fre（本文）	24.37
L_spa+0.15×L_fre	24.08

Low-light image enhancement network guided by reflection prior map

反射先验图引导的低光图像增强网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 14

References 30

Related Articles 15

Recommended Articles

Metrics

方法	PSNR	方法	PSNR
变体1	26.61	变体4	26.40
变体2	26.61	变体5	26.76
变体3	26.66	本文方法	26.96

[1]	Qianhui XU, Ke NIU, Shunzhe ZHU, Lin SHI, Jun LI. GAB3D-SEVSN： enhanced video steganography model via invertible neural network [J]. Journal of Computer Applications, 2026, 46(2): 467-474.
[2]	Hu LUO, Mingshu ZHANG. Rumor detection method based on cross-modal attention mechanism and contrastive learning [J]. Journal of Computer Applications, 2026, 46(2): 361-367.
[3]	Jinjiao LIN, Canshun ZHANG, Shuya CHEN, Tianxin WANG, Jian LIAN, Yonghui XU. Vehicle insurance fraud detection method based on improved graph attention network [J]. Journal of Computer Applications, 2026, 46(2): 437-444.
[4]	Junrui WU, Jiangchuan YANG, Haisheng YU, Sai ZOU, Wenyong WANG. Performance evaluation method for deterministic networks based on complex-enhanced attention graph neural network [J]. Journal of Computer Applications, 2026, 46(2): 505-517.
[5]	Ming LI, Mengqi WANG, Aili ZHANG, Hua REN, Yuqiang DOU. Image steganography method based on conditional generative adversarial networks and hybrid attention mechanism [J]. Journal of Computer Applications, 2026, 46(2): 475-484.
[6]	Yingjie MA, Jingying QIN, Geng ZHAO, Jing XIAO. Deep compressive sensing network for IoT images and its chaotic encryption protection method [J]. Journal of Computer Applications, 2026, 46(1): 144-151.
[7]	Yanan LI, Mengyang GUO, Guojun DENG, Yunfeng CHEN, Jianji REN, Yongliang YUAN. Method for life prediction of parallel branching engine based on multi-modal fusion features [J]. Journal of Computer Applications, 2026, 46(1): 305-313.
[8]	Lifang WANG, Wenjing REN, Xiaodong GUO, Rongguo ZHANG, Lihua HU. Trident generative adversarial network for low-dose CT image denoising [J]. Journal of Computer Applications, 2026, 46(1): 270-279.
[9]	Weigang LI, Jiale SHAO, Zhiqiang TIAN. Point cloud classification and segmentation network based on dual attention mechanism and multi-scale fusion [J]. Journal of Computer Applications, 2025, 45(9): 3003-3010.
[10]	Xiang WANG, Zhixiang CHEN, Guojun MAO. Multivariate time series prediction method combining local and global correlation [J]. Journal of Computer Applications, 2025, 45(9): 2806-2816.
[11]	Xuejin WANG, Leilei HUANG, Zhenhui ZHONG. Noise and semantic prior guided low-light image enhancement algorithm [J]. Journal of Computer Applications, 2025, 45(9): 2966-2974.
[12]	Jinggang LYU, Shaorui PENG, Shuo GAO, Jin ZHOU. Speech enhancement network driven by complex frequency attention and multi-scale frequency enhancement [J]. Journal of Computer Applications, 2025, 45(9): 2957-2965.
[13]	Chao JING, Yutao QUAN, Yan CHEN. Improved multi-layer perceptron and attention model-based power consumption prediction algorithm [J]. Journal of Computer Applications, 2025, 45(8): 2646-2655.
[14]	Jinhao LIN, Chuan LUO, Tianrui LI, Hongmei CHEN. Thoracic disease classification method based on cross-scale attention network [J]. Journal of Computer Applications, 2025, 45(8): 2712-2719.
[15]	Jin ZHOU, Yuzhi LI, Xu ZHANG, Shuo GAO, Li ZHANG, Jiachuan SHENG. Modulation recognition network for complex electromagnetic environments [J]. Journal of Computer Applications, 2025, 45(8): 2672-2682.