光照先验引导的低照度图像增强方法

张凯兵; 任可心; 李英健; 孟雅蕾; 李鹏芳

doi:10.16152/j.cnki.xdxbzr.2026-03-007

您当前的位置：

首页 >

文章列表页 >

光照先验引导的低照度图像增强方法

深度学习视觉处理与分析 | 更新时间：2026-07-07

- 光照先验引导的低照度图像增强方法
- Illumination prior guided low-light image enhancement method
- 西北大学学报（自然科学版） 2026年56卷第3期页码：538-550
- 作者机构：
  
  1.西安工程大学　计算机科学学院，陕西　西安　710048
  2.陕西省服装设计智能化重点实验室，陕西　西安　710048
- 作者简介：
  
  张凯兵，男，博士，教授，博士生导师，从事机器学习、图像超分辨重建等研究，zhangkaibing@xpu.edu.cn。
- 基金信息：
  
  国家自然科学基金(61971339);陕西省重点产业创新链项目(2024GX-ZDCYL-02-07)
- DOI：10.16152/j.cnki.xdxbzr.2026-03-007
  中图分类号： TP391.41
- 收稿：2025-12-26，
  
  修回：2026-01-08，
  
  纸质出版：2026-06-25
- 稿件说明：
移动端阅览
张凯兵, 任可心, 李英健, 等. 光照先验引导的低照度图像增强方法[J]. 西北大学学报（自然科学版）, 2026,56(3):538-550.

ZHANG Kaibing, REN Kexin, LI Yingjian, et al. Illumination prior guided low-light image enhancement method[J]. Journal of Northwest University (Natural Science Edition), 2026, 56(3): 538-550.
张凯兵, 任可心, 李英健, 等. 光照先验引导的低照度图像增强方法[J]. 西北大学学报（自然科学版）, 2026,56(3):538-550. DOI： 10.16152/j.cnki.xdxbzr.2026-03-007.

ZHANG Kaibing, REN Kexin, LI Yingjian, et al. Illumination prior guided low-light image enhancement method[J]. Journal of Northwest University (Natural Science Edition), 2026, 56(3): 538-550. DOI： 10.16152/j.cnki.xdxbzr.2026-03-007.

摘要

针对现有低照度图像增强方法存在光照适应性差、全局结构与局部细节难以平衡的问题，提出一种光照先验引导的低照度图像增强方法（light-prior driven low-light enhancement network，LPD-Net）。首先，基于Retinex理论构建光照先验估计模块（illumination prior module，IPM），通过引入同态滤波先验生成光照归一化特征，实现对非均匀光照的全局校正。其次，构建基于编码器解码器架构的Mamba优化增强模块（Mamba enhanced module，MEM），采用分层建模策略，构建由局部窗口Mamba与全局Mamba组成的多尺度路径，实现局部细节增强与全局亮度恢复。在此基础上，引入双流融合色彩调制模块（dual-stream fusioncolor modulation，DFCM），通过双向卷积块获取横纵空间依赖关系，并结合轻量注意力机制实现全局局部特征的融合。同时，利用通道门控与色彩调制机制自适应恢复色彩信息与局部图像细节。最后，通过引入总变差正则化对光照图进行约束，防止因过度增强而导致的图像结构一致性问题。实验结果表明，本文方法在LOLv1数据集上取得了24.370 dB的PSNR与0.853的SSIM，性能明显优于现有方法。该方法在光照校正、噪声抑制和细节保持等方面均展现出显著优势，同时对实际低照度场景也具有良好的泛化能力。

Abstract

Aiming at the shortcomings of existing low-light image enhancement methods

such as poor illumination adaptability and difficulty in balancing global structure and local details

this paper proposes a light-prior driven low-light enhancement network (LPD-Net). First

an illumination prior module (IPM) is constructed based on Retinex theory

which generates illumination-normalized features by introducing homomorphic filtering priors to achieve global correction of non-uniform illumination. Second

a Mamba enhanced module (MEM) based on an encoder-decoder architecture is designed

employing a hierarchical modeling strategy and constructing a multi-scale pathway consisting of local window Mamba and global Mamba to achieve local detail enhancement and global brightness restoration. On this basis

a dual-stream fusion color modulation (DFCM) module is introduced

which captures horizontal and vertical spatial dependencies through bidirectional convolutional blocks and integrates global-local features using a lightweight attention mechanism. Simultaneously

channel gating and color modulation mechanisms are applied to adaptively restore color information and fine image details. Finally

total variation regularization is introduced to constrain the illumination map

preventing structural inconsistency caused by over-enhancement. Experimental results show that the proposed method achieves 24.370 dB PSNR and 0.853 SSIM on the LOLv1 dataset

significantly outperforming existing methods. The method demonstrates notable advantages in illumination correction

noise suppression

and detail preservation

while also exhibiting strong generalization capability for real-world low-light scenarios.

关键词

Keywords

references

董良振，田建艳，杨胜强，等 . 基于光照校正和图像融合的零件表面图像增强 [J ] . 计算机工程， 2024 , 50 ( 6 ): 245 - 254 .

DONG L Z , TIAN J Y , YANG S Q , et al . Image enhancement of part surface based on illumination correction and image fusion [J ] . Computer Engineering , 2024 , 50 ( 6 ): 245 - 254 .

CHEN T , LYU Q C , BAI L , et al . LighTDiff: Surgical endoscopic image low-light enhancement with T-diffusion [C ] // Medical Image Computing and Computer-Assisted Intervention - MICCAI 2024 . Cham : Springer , 2024 : 369 - 379 .

杨允权，刘国宁，孙克文，等 . 基于神经网络的飞行器空中侦查时低照度图像增强系统 [J ] . 机械设计与制造， 2023 ( 9 ): 163 - 168 .

YANG Y Q , LIU G N , SUN K W , et al . Neural network based low-light aerial image enhancement system for aerial surveillance by aircraft [J ] . Machinery Design & Manufacture , 2023 ( 9 ): 163 - 168 .

PIZER S M , AMBURN E P , AUSTIN J D , et al . A-daptive histogram equalization and its variations [J ] . Computer Vision, Graphics, and Image Processing , 1987 , 39 ( 3 ): 355 - 368 .

李学明 . 基于Retinex理论的图像增强算法 [J ] . 计算机应用研究， 2005 , 22 ( 2 ): 235 - 237 .

LI X M . Image enhancement algorithm based on Retinex theory [J ] . Application Research of Computers , 2005 , 22 ( 2 ): 235 - 237 .

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 .

WEI C , WANG W J , YANG W H , et al . Deep retinex decomposition for low-light enhancement [EB/OL ] . ( 2018-08-14 )[ 2025-12-05 ] . https://arxiv.org/abs/1808.04560 https://arxiv.org/abs/1808.04560 .

王萍，孙振明 . 多级分解的Retinex低照度图像增强算法 [J ] . 计算机应用研究， 2020 , 37 ( 4 ): 1204 - 1209 .

WANG P , SUN Z M . Multi-level decomposition Retinex low-light image enhancement algorithm [J ] . Application Research of Computers , 2020 , 37 ( 4 ): 1204 - 1209 .

LIU R S , MA L , ZHANG J A , et al . Retinex-inspired unrolling with cooperative prior architecture search for low-light image enhancement [C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Nashville : IEEE , 2021 : 10556 - 10565 .

MA L , MA T Y , LIU R S , et al . Toward fast, flexible, and robust low-light image enhancement [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition . New Orleans : IEEE , 2022 : 5627 - 5636 .

XU X G , WANG R X , FU C W , et al . SNR-aware low-light image enhancement [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition . New Orleans : IEEE , 2022 : 17693 - 17703 .

PEI X , HUANG Y D , SU W J , et al . FFTFormer: A spatial-frequency noise aware CNN-Transformer for lowlight image enhancement [J ] . Knowledge-Based Systems , 2025 , 314 : 113055 .

WEN Y J , XU P , LI Z H , et al . An illumination-guided dual attention vision transformer for low-light image enhancement [J ] . Pattern Recognition , 2025 , 158 : 111033 .

徐文晨，樊佳庆，宋慧慧 . 基于上下文Transformer的低光照图像增强网络 [J ] . 计算机与数字工程， 2023 , 51 ( 1 ): 237 - 244 .

XU W C , FAN J Q , SONG H H . Low lighting image enhancement network based on contextual Transformer [J ] . Computer and Digital Engineering , 2023 , 51 ( 1 ): 237 - 244 .

ZOU W B , GAO H X , YANG W P , et al . Wave-mamba: Wavelet state space model for ultra-high-definition low-light image enhancement [C ] // Proceedings of the 32nd ACM International Conference on Multimedia . Melbourne : ACM , 2024 : 1534 - 1543 .

BAI J S , YIN Y H , HE Q Y , et al . Retinexmamba: Retinex-based mamba for low-light image enhancement [M ] // International Conference on Neural Information Processing . Singapore : Springer Nature Singapore , 2025 : 427 - 442 .

CHEN K , ZHANG K B , PANG F F , et al . RMANet: Refined-mixed attention network for progressive low-light image enhancement [J ] . Signal Processing , 2025 , 227 : 109689 .

WANG H K , YAN X Y , HOU X S , et al . Extracting noise and darkness: Low-light image enhancement via dual prior guidance [J ] . IEEE Transactions on Circuits and Systems for Video Technology , 2025 , 35 ( 2 ): 1700 - 1714 .

朱松浩，赵芝芸，王梦灵 . 基于轻量残差与亮度感知动态特征融合的低光图像增强网络 [J/OL ] . 计算机应用 .（ 2025-09-28 ）[ 2025-12-05 ] . https://link.cnki.net/urlid/51.1307.TP.20250925.1330.034 https://link.cnki.net/urlid/51.1307.TP.20250925.1330.034 .

ZHU S H , ZHAO Z Y , WANG M L . Lightweight residual and brightness-aware dynamic feature fusion for low-light image enhancement [J/OL ] . Journal of Computer Applications .( 2025-09-28 )[ 2025-12-05 ] . https://link.cnki.net/urlid/51.1307.TP.20250925.1330.034 https://link.cnki.net/urlid/51.1307.TP.20250925.1330.034 .

江泽涛，李堃，黄钦阳，等 . 一种基于自校准知识蒸馏的低照度图像增强方法 [J/OL ] . 计算机辅助设计与图形学学报 .（ 2025-04-03 ）[ 2025-12-05 ] . https://link.cnki.net/urlid/11.2925.tp.20250403.1427.004 https://link.cnki.net/urlid/11.2925.tp.20250403.1427.004 .

JIANG Z T , LI K , HUANG Q Y , et al . Self-calibrated knowledge distillation for low-light image enhancement [J/OL ] . Journal of Computer-Aided Design & Computer Graphics . ( 2025-04-03 ) [ 2025-12-05 ] . https://link.cnki.net/urlid/11.2925.tp.20250403.1427.004 https://link.cnki.net/urlid/11.2925.tp.20250403.1427.004 .

WANG H K , YAN X Y , HOU X S , et al . Division gets better: Learning brightness-aware and detail-sensitive representations for low-light image enhancement [J ] . Knowledge-Based Systems , 2024 , 299 : 111958 .

LI Y J , ZHANG K B , ZHOU X , et al . Rethinking progressive low-light image enhancement: A frequency-aware tripartite multi-scale network [J ] . Neural Networks , 2026 , 196 : 108351 .

张凯兵，张智启，孟雅蕾 . 语义感知下的多尺度U型Swin Transformer人脸超分辨方法 [J ] . 西安工程大学学报， 2025 , 39 ( 5 ): 135 - 145 .

ZHANG K B , ZHANG Z Q , MENG Y L . Semantic-aware multi-scale U-shaped Swin Transformer for face super-resolution [J ] . Journal of Xi’an Polytechnic University , 2025 , 39 ( 5 ): 135 - 145 .

王晓华，侯佳辉，张凯兵，等 . 渐进式深度网络下盲运动图像去模糊方法 [J ] . 西安工程大学学报， 2023 , 37 ( 3 ): 74 - 82 .

WANG X H , HOU J H , ZHANG K B , et al . Progressive deep network for blind motion image deblurring [J ] . Journal of Xi’an Polytechnic University , 2023 , 37 ( 3 ): 74 - 82 .

LYU F , LU F , WU J , et al . MBLLEN: Low-light image/video enhancement using CNNs [C ] // Proceedings of the British Machine Vision Conference . Newcastle : BMVA Press , 2018 , 220 ( 1 ): 4 .

ZHANG Y H , ZHANG J W , GUO X J . Kindling the darkness: A practical low-light image enhancer [C ] // Proceedings of the 27th ACM International Conference on Multimedia . Nice : ACM , 2019 : 1632 - 1640 .

ZHANG Y H , GUO X J , MA J Y , et al . Beyond brightening low-light images [J ] . International Journal of Computer Vision , 2021 , 129 ( 4 ): 1013 - 1037 .

GUO C L , LI C Y , GUO J C , et al . Zero-reference deep curve estimation for low-light image enhancement [C ] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Seattle : IEEE , 2020 : 1777 - 1786 .

JIANG Y F , GONG X Y , LIU D , et al . Enlight-enGAN: Deep light enhancement without paired supervision [J ] . IEEE Transactions on Image Processing , 2021 , 30 : 2340 - 2349 .

ZAMIR S W , ARORA A , KHAN S , et al . Learning enriched features for fast image restoration and enhancement [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023 , 45 ( 2 ): 1934 - 1948 .

CAI Y H , BIAN H , LIN J , et al . Retinexformer: Onestage retinex-based transformer for low-light image enhancement [C ] // 2023 IEEE/CVF International Conference on Computer Vision . Paris : IEEE , 2023 : 12470 - 12479 .

WANG T , ZHANG K H , SHEN T R , et al . Ultra-high-definition low-light image enhancement: A benchmark and transformer-based method [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2023 , 37 ( 3 ): 2654 - 2662 .

ZHANG K B , YUAN C , LI J , et al . Multi-branch and progressive network for low-light image enhancement [J ] . IEEE Transactions on Image Processing , 2023 , 32 : 2295 - 2308 .

WANG H K , HOU X S , LI J T , et al . Multi-scale Retinex unfolding network for low-light image enhancement [J ] . IEEE Transactions on Multimedia , 2025 , 27 : 5709 - 5721 .

ZHANG J M , JIANG J , FENG Z J , et al . Learning spatial-channel feature refiners via wavelet-based linear mixed basis for low-light image enhancement [J ] . Knowledge-Based Systems , 2025 , 330 : 114567 .

CUI Z T , LI K C , GU L , et al . You only need 90K parameters to adapt light: A light weight transformer for image enhancement and exposure correction [EB/OL ] . ( 2022-05-27 )[ 2025-12-05 ] . https://arxiv.org/abs/2205.14871 https://arxiv.org/abs/2205.14871 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

多分知识结构的基与交式生成组

基于双域聚合Transformer的高光谱图像超分辨重建方法

遗址公园与城市空间关联耦合特征及影响因素：以西安市中心城区为例

鄂尔多斯盆地延长探区东部马五1亚段沉积相特征及展布规律

西北地区民宿空间分布特征及其影响因素研究