
浏览全部资源
扫码关注微信
1.太原师范学院 计算机科学与技术学院, 山西 晋中 030619
2.智能优化计算与区块链技术山西省重点实验室, 山西 晋中 030619
3.西北师范大学 物理与电子工程学院, 甘肃 兰州 730070
杨潞霞,女,博士,教授,2014年于中北大学获得博士学位,主要从事智能信息处理方面的研究。E-mail:ylxfyz328@163.com
张红瑞,女,博士,讲师,2023年于华中师范大学获得博士学位,主要从事智能交通系统下机器视觉的研究。E-mail:zhanghongrui@tynu.edu.cn
收稿:2026-03-03,
修回:2026-03-20,
纸质出版:2026-04-05
移动端阅览
杨潞霞, 郭鑫, 张红瑞, 等. 基于区域对比与边界感知的阴影去除网络[J]. 液晶与显示, 2026,41(4):578-591.
YANG Luxia, GUO Xin, ZHANG Hongrui, et al. Shadow removal network based on regional contrast and boundary perception[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(4): 578-591.
杨潞霞, 郭鑫, 张红瑞, 等. 基于区域对比与边界感知的阴影去除网络[J]. 液晶与显示, 2026,41(4):578-591. DOI: 10.37188/CJLCD.2026-0025. CSTR: 32172.14.CJLCD.2026-0025.
YANG Luxia, GUO Xin, ZHANG Hongrui, et al. Shadow removal network based on regional contrast and boundary perception[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(4): 578-591. DOI: 10.37188/CJLCD.2026-0025. CSTR: 32172.14.CJLCD.2026-0025.
针对现有阴影去除方法在全局对比关系建模不足、区域差异化处理能力有限以及易产生阴影残留与边界伪影等问题,提出一种基于区域对比与边界感知的阴影去除网络。首先,设计多尺度注意力机制,利用互补空间表征以及关键区域增强提高图像结构一致性;其次,构建阴影感知模块,对阴影区域与非阴影区域采用差异化特征处理策略,提升阴影边界及其过渡区域的连续性;然后,提出视觉对比注意力机制,在全局尺度上提高对跨区域对比关系的界定能力;最后,设计联合损失函数,通过动态权重调节策略,引导模型实现像素精度、结构一致性与色彩稳定性的协同优化。实验结果表明,在ISTD数据集上与8种代表性方法比较PSNR平均提高了1.57 dB,在SRD数据集上与10种代表方法比较PSNR平均提高了2.03 dB,同时,本文方法以12.3M的参数量,实现了重建质量与模型复杂度的有效平衡。该方法显著降低了像素级亮度误差并有效抑制了阴影残留与边界伪影,实现更加自然、可靠的阴影去除效果。
To address the limitations of existing shadow removal methods, such as insufficient modeling of global contrast relationships, limited regional differentiation processing capabilities, and the tendency to produce shadow residuals and boundary artifacts, a shadow removal network based on regional contrast and boundary perception is proposed.Firstly, a multi-scale attention mechanism is designed to enhance the consistency of image structure by utilizing complementary spatial representations and key region enhancement. Secondly, a shadow perception module is constructed to apply differentiated feature processing strategies to shadow and non-shadow regions, improving the continuity of shadow boundaries and their transition areas. Then, a visual contrast attention mechanism is proposed to enhance the ability to define cross-regional contrast relationships at the global scale. Finally, a joint loss function is designed, with a dynamic weight adjustment strategy to guide the model to achieve a coordinated optimization of pixel accuracy, structural consistency, and color stability.The experimental results show that, compared with 8 representative methods on the ISTD dataset, the PSNR is improved by an average of 1.57 dB, and compared with 10 representative methods on the SRD dataset, the PSNR is improved by an average of 2.03 dB. Meanwhile, with 12.3M parameters, the proposed method achieves an effective balance between reconstruction quality and model complexity.This method significantly reduces pixel-level brightness errors and effectively suppresses shadow residuals and boundary artifacts, achieving a more natural and reliable shadow removal effect.
HONG X Y , XIE J , SHENG J , et al . Single-view neural illumination estimation and editing for dynamic light field display [J]. Light: Science & Applications , 2026 , 15 ( 1 ): 147 .
LIU Z Y , GAO H , MA T G , et al . Broadband spin and angle co-multiplexed waveguide-based metasurface for six-channel crosstalk-free holographic projection [J]. eLight , 2024 , 4 ( 1 ): 7 .
WANG X , YAO S Y , TANG Y , et al . Shadow-aware decomposed transformer network for shadow detection and removal [J]. Pattern Recognition , 2024 , 156 : 110771 .
王笳辉 , 彭光灵 , 段亮 , 等 . 面向文本识别的小样本阴影消除方法 [J]. 计算机科学 , 2024 , 51 ( 9 ): 147 - 154 .
WANG J H , PENG G L , DUAN L , et al . Few-shot shadow removal method for text recognition [J]. Computer Science , 2024 , 51 ( 9 ): 147 - 154 . (in Chinese)
JIE L P , ZHANG H . RMLANet: random multi-level attention network for shadow detection and removal [J]. IEEE Transactions on Circuits and Systems for Video Technology , 2023 , 33 ( 12 ): 7819 - 7831 .
邸敬 , 梁婵 , 刘冀钊 , 等 . 基于跨域交互注意力和对比学习引导的红外与可见光图像融合 [J]. 中国光学(中英文) , 2025 , 18 ( 2 ): 317 - 332 .
DI J , LIANG C , LIU J Z , et al . Infrared and visible image fusion guided by cross-domain interactive attention and contrastive learning [J]. Chinese Optics , 2025 , 18 ( 2 ): 317 - 332 . (in Chinese)
DONG W , ZHOU H , MOUSAVI S A , et al . Retinex-guided histogram transformer for mask-Free shadow removal [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops . Nashville : IEEE , 2025 : 1462 - 1472 .
高世严 , 柳杰 , 陈文艺 , 等 . 基于深度学习的单目标跟踪算法研究进展 [J]. 液晶与显示 , 2025 , 40 ( 8 ): 1202 - 1218 .
GAO S Y , LIU J , CHEN W Y , et al . Review of single object tracking algorithm based on deep learning [J]. Chinese Journal of Liquid Crystals and Displays , 2025 , 40 ( 8 ): 1202 - 1218 . (in Chinese)
GUO R Q , DAI Q Y , HOIEM D . Paired regions for shadow detection and removal [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2013 , 35 ( 12 ): 2956 - 2967 .
FINLAYSON G D , DREW M S , LU C . Entropy minimization for shadow removal [J]. International Journal of Computer Vision , 2009 , 85 ( 1 ): 35 - 57 .
FINLAYSON G D , HORDLEY S D , LU C , et al . On the removal of shadows from images [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2006 , 28 ( 1 ): 59 - 68 .
QU L Q , TIAN J D , HE S F , et al . DeshadowNet: a multi-context embedding deep network for shadow removal [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Honolulu : IEEE , 2017 : 2308 - 2316 .
FAN H , HAN M , LI J J . Image shadow removal using end-to-end deep convolutional neural networks [J]. Applied Sciences , 2019 , 9 ( 5 ): 1009 .
GUO L Q , HUANG S Y , LIU D , et al . ShadowFormer: global context helps shadow removal [C]// Proceedings of the 37th AAAI Conference on Artificial Intelligence . Washington : AAAI Press , 2023 : 710 - 718 .
WANG D H , WANG J H , HE N , et al . Enhancing unsupervised shadow removal via multi-intensity shadow generation and diffusion modeling [J]. The Visual Computer , 2025 , 41 ( 8 ): 5461 - 5476 .
WANG D H , WANG J H , HE N , et al . Image shadow removal algorithm based on MaskGuideAttention and mask-aware enhancement module [J]. Multimedia Systems , 2025 , 31 ( 3 ): 224 .
黄颖 , 程彬 , 房少杰 , 等 . 用于阴影去除的小波非均匀扩散模型 [J]. 中国图象图形学报 , 2025 , 30 ( 1 ): 66 - 82 .
HUANG Y , CHENG B , FANG S J , et al . Shadow removal with wavelet-based non-uniform diffusion model [J]. Journal of Image and Graphics , 2025 , 30 ( 1 ): 66 - 82 . (in Chinese)
黄璞 , 苏畅 , 杨章静 , 等 . 双通道噪声抑制网络及在阴影去除中的应用 [J]. 小型微型计算机系统 , 2025 , 46 ( 10 ): 2431 - 2439 .
HUANG P , SU C , YANG Z J , et al . Dual-channel noise suppression network and its application in shadow removal [J]. Journal of Chinese Computer Systems , 2025 , 46 ( 10 ): 2431 - 2439 . (in Chinese)
黄颖 , 雷秀杰 , 姜茂 , 等 . 基于特征分离和差异映射的阴影去除算法 [J]. 计算机辅助设计与图形学学报 , 2025 , 37 ( 8 ): 1383 - 1393 .
HUANG Y , LEI X J , JIANG M , et al . Feature separation and differential mapping based shadow removal algorithm [J]. Journal of Computer-Aided Design & Computer Graphics , 2025 , 37 ( 8 ): 1383 - 1393 . (in Chinese)
LI X J , ZHAO Y , CHEN Y , et al . Towards non-uniform shadow removal using shadow adaptive normalization [J]. Digital Signal Processing , 2025 , 166 : 105364 .
SHAO Y M , ZHANG Z J , YANG M M . Symmetry-guided AB-dynamic feature refinement network for weakly supervised shadow removal [J]. Symmetry , 2026 , 18 ( 2 ): 330 .
LI Y , HE J D , ZOU D P , et al . Image shadow detection and removal via illumination estimation and chromaticity guidance [J]. IEEE Transactions on Automation Science and Engineering , 2026 , 23 : 3661 - 3672 .
LIU D Y , ZHANG J P , WU Y H , et al . A shadow detection algorithm based on multiscale spatial attention mechanism for aerial remote sensing images [J]. IEEE Geoscience and Remote Sensing Letters , 2022 , 19 : 6003905 .
WANG X Q , DAI J L , CHEN B , et al . A dynamic alignment and illumination-aware convolution for shadow removal [J]. IET Image Processing , 2024 , 18 ( 3 ): 785 - 795 .
LIU Z , LIN Y T , CAO Y , et al . Swin transformer: hierarchical vision transformer using shifted windows [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision . Montreal, QC, Canada : IEEE , 2021 : 9992 - 10002 .
WANG J F , LI X , YANG J . Stacked conditional generative adversarial networks for jointly learning shadow detection and shadow removal [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City : IEEE , 2018 : 1788 - 1797 .
CUN X D , PUN C M , SHI C . Towards ghost-free shadow removal via dual hierarchical aggregation network and shadow matting GAN [C]// Proceedings of the 34th AAAI Conference on Artificial Intelligence . New York : AAAI Press , 2020 : 10680 - 10687 .
HU X W , ZHU L , FU C W , et al . Direction-aware spatial context features for shadow detection [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City : IEEE , 2018 : 7454 - 7462 .
FU L , ZHOU C Q , GUO Q , et al . Auto-exposure fusion for single-image shadow removal [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Nashville : IEEE , 2021 : 10566 - 10575 .
LIANG J J , ZHANG B W , DENG J W , et al . Towards hard and soft shadow removal via dual-branch separation network and vision transformer [C]// Proceedings of the International Conference on Machine Learning and Cybernetics . Miyazaki : IEEE , 2024 : 519 - 526 .
WAN J , YIN H , WU Z Y , et al . CRFormer: a cross-region transformer for shadow removal [J]. Image and Vision Computing , 2024 , 151 : 105273 .
GUO M Q , CAO R , HUANG Y , et al . Image shadow removal based on pseudo-illuminated region and log domain transformation [J]. Journal of Spatial Science , 2025 , 70 ( 2 ): 253 - 273 .
LIU Z H , YIN H , WU X Y , et al . From shadow generation to shadow removal [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Nashville : IEEE , 2021 : 4925 - 4934 .
ZHU Y R , HUANG J , FU X Y , et al . Bijective mapping network for shadow removal [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . New Orleans : IEEE , 2022 : 5617 - 5626 .
JIN Y Y , YE W , YANG W H , et al . DeS3: adaptive attention-driven self and soft shadow removal using ViT similarity [C]// Proceedings of the 38th AAAI Conference on Artificial Intelligence . Vancouver : AAAI Press , 2024 : 2634 - 2642 .
0
浏览量
417
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621