并行多尺度特征递归学习的低照度图像增强

doi:10.3778/j.issn.1002-8331.2407-0185

摘要/Abstract

摘要： 针对在低照度条件下拍摄的图像存在的如细节和纹理信息模糊或缺失、色彩失真、噪声污染严重等退化问题，提出基于并行多尺度特征递归学习（parallel multi-scale feature recursive learning，PMRL）的图像增强方法。通过从粗到细的递归学习策略从退化的输入图像中逐步学习恢复映射，使照度更接近于真实光照；提出一种表示融合机制，使不同阶段之间的特征可以进行信息交换，以抑制噪声和伪影；引入捕获上下文信息的双重注意力表示特征提取模块，得到更丰富的图像细节信息；根据灰度世界颜色恒定性假设，提出半监督的感知损失来自适应地提高增强效果，保持颜色分布的一致性。在公开数据集LOLv1和LOLv2上的实验结果表明，相较于ZeroDCE++、DRBN和MIRNet等网络，峰值信噪比提高了8.18%~55.59%，结构相似性提高了5.42%~53.44%，验证了该方法的优越性。

关键词: 低照度图像增强, 递归学习, 多尺度特征融合, 注意力机制

Abstract: To solve the degradation problems such as fuzzy or missing details and texture information, color distortion and serious noise pollution in images taken under low-light environment, an image enhancement method based on parallel multi-scale feature recursive learning (PMRL) is proposed. Through the coarse-to-fine recursive learning strategy, the restoration map is gradually learned from the degraded input image to make the illumination closer to the real illumination. In addition, a representation fusion mechanism is developed to exchange information between different stages to suppress noise and artifacts. The dual attention representation feature extraction module that captures context information is introduced to obtain more image detail information. Based on the color constancy assumption in the grayscale world, the semi-supervised perceptual loss is proposed to adaptively improve the image enhancement effect and maintain the consistency of color distribution. Experimental results on public datasets LOLv1 and LOLv2 show that compared with ZeroDCE++, DRBN, MIRNet and other networks, the peak signal-to-noise ratio is increased by 8.18%~55.59%, and the structural similarity is increased by 5.42%~53.44%, which verifies the superiority of the proposed method.

Key words: low-light image enhancement, recursive learning, multi-scale feature fusion, attention mechanism

王璐雪, 王晓霞, 李翔, 陈晓. 并行多尺度特征递归学习的低照度图像增强[J]. 计算机工程与应用, 2025, 61(16): 265-271.

WANG Luxue, WANG Xiaoxia, LI Xiang, CHEN Xiao. Parallel Multi-Scale Feature Recursive Learning for Low-Light Image Enhancement[J]. Computer Engineering and Applications, 2025, 61(16): 265-271.

参考文献

[1] CHO S J, JI S W, HONG J P, et al. Rethinking coarse-to-fine approach in single image deblurring[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2021: 4621-4630.
[2] XU X G, WANG R X, 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.
[3] ZHANG Z, ZHENG H, HONG R C, et al. Deep color consistent network for low-light image enhancement[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 1889-1898.
[4] LEE C, LEE C, KIM C S. Contrast enhancement based on layered difference representation of 2D histograms[J]. IEEE Transactions on Image Processing, 2013, 22(12): 5372-5384.
[5] REN X T, YANG W H, CHENG W H, et al. LR3M: robust low-light enhancement via low-rank regularized retinex model[J]. IEEE Transactions on Image Processing, 2020, 29: 5862-5876.
[6] HAO S J, HAN X, GUO Y R, et al. Low-light image enhancement with semi-decoupled decomposition[J]. IEEE Transactions on Multimedia, 2020, 22(12): 3025-3038.
[7] 马龙, 马腾宇, 刘日升. 低光照图像增强算法综述[J]. 中国图象图形学报, 2022, 27(5): 1392-1409.
MA L, MA T Y, LIU R S. The review of low-light image enhancement[J]. Journal of Image and Graphics, 2022, 27(5): 1392-1409.
[8] 朱凯, 李理, 张彤, 等. 视觉Transformer在低级视觉领域的研究综述[J]. 计算机工程与应用, 2024, 60(4): 39-56.
ZHU K, LI L, ZHANG T, et al. Survey of vision Transformer in low-level computer vision[J]. Computer Engineering and Applications, 2024, 60(4): 39-56.
[9] 陈辉, 王硕, 许家昌, 等. 基于多尺度特征融合生成对抗网络的水下图像增强[J]. 计算机工程与应用, 2023, 59(21): 231-241.
CHEN H, WANG S, XU J C, et al. Underwater image enhancement based on generate adversarial network with multiscale feature fusion[J]. Computer Engineering and Applications, 2023, 59(21): 231-241.
[10] WANG X Z, CHEN K Q, WANG Z X, et al. PMSNet: parallel multi-scale network for accurate low-light light-field image enhancement[J]. IEEE Transactions on Multimedia, 2023, 26: 2041-2055.
[11] LEE S, JANG D, KIM D S. Temporally averaged regression for semi-supervised low-light image enhancement[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway: IEEE, 2023: 4208-4217.
[12] OZCAN M, ERGEZER H, AYAZO?LU M. Flight mode on: a feather-light network for low-light image enhancement[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 4225-4234.
[13] LI J J, FENG X M, HUA Z. Low-light image enhancement via progressive-recursive network[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 31(11): 4227-4240.
[14] 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.
[15] CAI Y H, 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.
[16] 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.
[17] ZHANG Y L, TIAN Y P, KONG Y, et al. Residual dense network for image restoration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 43(7): 2480-2495.
[18] LI X, WANG W H, HU X L, et al. Selective kernel networks[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 510-519.
[19] GUO C L, LI C Y, GUO J C, 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] LI C Y, GUO C L, LOY C C. Learning to enhance low-light image via zero-reference deep curve estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(8): 4225-4238.
[21] LIU R S, MA L, ZHANG J A, 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.
[22] YANG W H, WANG S Q, FANG Y M, et al. From fidelity to perceptual quality: a semi-supervised approach for low-light image enhancement[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 3060-3069.