A novel image enhancement method based on image decomposition and deep neural networks
•We propose a novel image decomposition-based optimization model for low-light image enhancement, by using the total variation and multi-scale convolutional sparse coding to precisely represent illumination and reflectance layers.•We integrate deep unfolding networks into framework of the proposed o...
Saved in:
| Published in: | Pattern recognition Vol. 172; p. 112371 |
|---|---|
| Main Authors: | , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
01.04.2026
|
| Subjects: | |
| ISSN: | 0031-3203 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | •We propose a novel image decomposition-based optimization model for low-light image enhancement, by using the total variation and multi-scale convolutional sparse coding to precisely represent illumination and reflectance layers.•We integrate deep unfolding networks into framework of the proposed optimization algorithm.•The deep unfolding network automatically estimates the priors from training samples.•Experimental results demonstrate that the proposed method outperforms state-of-the-art image enhancement methods in terms of visual quality and robutsness.
Image decomposition and deep learning are active research areas in computer vision tasks, such as cartoon texture decomposition, low-light image enhancement, rain streak removal, image recovery, etc. This paper proposes a novel low-light image enhancement method by joining image decomposition and deep neural network techniques. We introduce a new image decomposition-based optimization model by incorporating the Tikhonov regularization and multi-scale convolutional sparse coding (MSCSC) to enhance image visual effects. To enhance robustness performance, we introduce a noise-free image decomposition error term to effectively suppress noise in low-light images. To effectively implement the proposed method, we incorporate a deep-unfolding neural network and an adaptive denoiser into the alternating direction method of multipliers (ADMM) framework. Since the deep unfolding network can effectively simulate the optimization algorithm process, the interpretability of the network model is increased. Moreover, through end-to-end training, we can automatically estimate the two priors and parameter settings from training samples. Finally, qualitative and quantitative experiments demonstrate that the proposed method outperforms state-of-the-art image enhancement methods in terms of visual quality and robustness. The source code is available at https://github.com/cassiopeia-yxx/LLIE. |
|---|---|
| AbstractList | •We propose a novel image decomposition-based optimization model for low-light image enhancement, by using the total variation and multi-scale convolutional sparse coding to precisely represent illumination and reflectance layers.•We integrate deep unfolding networks into framework of the proposed optimization algorithm.•The deep unfolding network automatically estimates the priors from training samples.•Experimental results demonstrate that the proposed method outperforms state-of-the-art image enhancement methods in terms of visual quality and robutsness.
Image decomposition and deep learning are active research areas in computer vision tasks, such as cartoon texture decomposition, low-light image enhancement, rain streak removal, image recovery, etc. This paper proposes a novel low-light image enhancement method by joining image decomposition and deep neural network techniques. We introduce a new image decomposition-based optimization model by incorporating the Tikhonov regularization and multi-scale convolutional sparse coding (MSCSC) to enhance image visual effects. To enhance robustness performance, we introduce a noise-free image decomposition error term to effectively suppress noise in low-light images. To effectively implement the proposed method, we incorporate a deep-unfolding neural network and an adaptive denoiser into the alternating direction method of multipliers (ADMM) framework. Since the deep unfolding network can effectively simulate the optimization algorithm process, the interpretability of the network model is increased. Moreover, through end-to-end training, we can automatically estimate the two priors and parameter settings from training samples. Finally, qualitative and quantitative experiments demonstrate that the proposed method outperforms state-of-the-art image enhancement methods in terms of visual quality and robustness. The source code is available at https://github.com/cassiopeia-yxx/LLIE. |
| ArticleNumber | 112371 |
| Author | Xia, Youshen Xiao, Yao |
| Author_xml | – sequence: 1 givenname: Yao surname: Xiao fullname: Xiao, Yao email: xiaoyao227192@163.com organization: College of Artificial Intelligence, Anhui University, HeFei, China – sequence: 2 givenname: Youshen surname: Xia fullname: Xia, Youshen email: ysxia2001@163.com organization: College of Artificial Intelligence, Anhui University, HeFei, China |
| BookMark | eNp9kM1OwzAQhH0oEm3hDTjkBRJs58fkglRV_FSqxAW4Wht707o0dmSbIt4eV-mZ00ijndHstyAz6ywScsdowShr7g_FCFG5XcEprwvGeCnYjMwpLVleclpek0UIB0qZYBWfk89VZt0Jj5kZYIcZ2j1YhQPamA0Y905nHQTUmbOXC43KDaMLJprkgdXJwTGz-O3hmCT-OP8VbshVD8eAtxddko_np_f1a759e9msV9tc8VrEXFeqfUAqmkYBdLwTjKm-bxl2CqpeQJ9mdw2ossdOCCjbigJtQNSsFjVvebkk1dSrvAvBYy9Hn3b6X8moPPOQBznxkGcecuKRYo9TDNO2k0EvgzKYHtfGo4pSO_N_wR_xC3Ad |
| Cites_doi | 10.1109/TCI.2024.3420942 10.1016/j.patcog.2024.111033 10.1109/TIP.2018.2839891 10.1016/j.knosys.2024.111779 10.1109/TCSVT.2022.3195996 10.1109/LSP.2022.3175096 10.1016/j.jvcir.2022.103712 10.1109/TIP.2021.3050850 10.1109/LSP.2012.2227726 10.1109/TCSVT.2021.3073371 10.1016/j.patcog.2022.109241 10.1016/j.patcog.2016.06.008 10.1109/TIP.2022.3189805 10.1016/j.patcog.2024.111076 10.1038/scientificamerican1277-108 10.1016/j.neucom.2022.12.043 10.1016/j.apm.2022.11.022 10.1109/TIP.2018.2810539 10.1109/TIP.2012.2214050 10.1109/TIP.2020.2974060 10.1073/pnas.80.16.5163 10.1109/TIP.2016.2639450 10.1109/TIP.2020.2984098 10.1023/A:1022314423998 10.1109/TIP.2006.888338 |
| ContentType | Journal Article |
| Copyright | 2025 Elsevier Ltd |
| Copyright_xml | – notice: 2025 Elsevier Ltd |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.patcog.2025.112371 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| ExternalDocumentID | 10_1016_j_patcog_2025_112371 S0031320325010325 |
| GroupedDBID | --K --M -D8 -DT -~X .DC .~1 0R~ 123 1B1 1RT 1~. 1~5 29O 4.4 457 4G. 53G 5VS 7-5 71M 8P~ 9JN AABNK AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AATTM AAXKI AAXUO AAYFN AAYWO ABBOA ABDPE ABEFU ABFNM ABFRF ABHFT ABJNI ABMAC ABWVN ABXDB ACBEA ACDAQ ACGFO ACGFS ACLOT ACNNM ACRLP ACRPL ACVFH ACZNC ADBBV ADCNI ADEZE ADJOM ADMUD ADMXK ADNMO ADTZH AEBSH AECPX AEFWE AEIPS AEKER AENEX AEUPX AFJKZ AFPUW AFTJW AGHFR AGQPQ AGUBO AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIGII AIIUN AIKHN AITUG AKBMS AKRWK AKYEP ALMA_UNASSIGNED_HOLDINGS AMRAJ ANKPU AOUOD APXCP ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFKBS EFLBG EJD EO8 EO9 EP2 EP3 F0J F5P FD6 FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HLZ HVGLF HZ~ H~9 IHE J1W JJJVA KOM KZ1 LG9 LMP LY1 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RNS ROL RPZ SBC SDF SDG SDP SDS SES SEW SPC SPCBC SST SSV SSZ T5K TN5 UNMZH VOH WUQ XJE XPP ZMT ZY4 ~G- ~HD 9DU AAYXX CITATION |
| ID | FETCH-LOGICAL-c257t-d4c98e0766caab2b711cff91ebca4f7af003b6ac3feb77a3940a06a7515752923 |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001567653900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0031-3203 |
| IngestDate | Sat Nov 29 07:00:38 EST 2025 Sat Oct 04 17:01:34 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Deep neural network Image decomposition Retinex theory Low-light image enhancement |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c257t-d4c98e0766caab2b711cff91ebca4f7af003b6ac3feb77a3940a06a7515752923 |
| ParticipantIDs | crossref_primary_10_1016_j_patcog_2025_112371 elsevier_sciencedirect_doi_10_1016_j_patcog_2025_112371 |
| PublicationCentury | 2000 |
| PublicationDate | April 2026 2026-04-00 |
| PublicationDateYYYYMMDD | 2026-04-01 |
| PublicationDate_xml | – month: 04 year: 2026 text: April 2026 |
| PublicationDecade | 2020 |
| PublicationTitle | Pattern recognition |
| PublicationYear | 2026 |
| Publisher | Elsevier Ltd |
| Publisher_xml | – name: Elsevier Ltd |
| References | Wei, Wang, Yang, Liu (bib0008) 2018 Fu, Yang, Tu, Huang, Ding, Ma (bib0009) 2023 Kimmel, Elad, Shaked, Keshet, Sobel (bib0016) 2003; 52 Wen, Xu, Li, ATO (bib0011) 2025; 158 Zhao, Xiong, Wang, Ou, Yu, Kuang (bib0018) 2022; 32 Liu, Ma, Ma, Fan, Luo (bib0033) 2023; 45 Xiao, Xia (bib0017) 2024 Ren, Yang, Cheng, Liu (bib0003) 2020; 29 Hai, Xuan, Yang, Hao, Zou, Lin, Han (bib0029) 2023; 90 Agaian, Silver, Panetta (bib0036) 2007; 16 Lin, Lu (bib0014) 2022; 31 Du, Zhao, Liu, You, Shi, Li, Xu (bib0015) 2023; 116 Lore, Akintayo, Sarkar (bib0007) 2017; 61 Guo, Li, Ling (bib0012) 2016; 26 Zhao, Zhang, Bai, Wang, Cui, Deng, Sun, Zhang, Liu, Xu (bib0026) 2023 Xu, Liu, Zhang, Wu, Zuo (bib0030) 2024; 295 Yang, Wang, Huang, Wang, Liu (bib0020) 2021; 30 Wu, Weng, Zhang, Wang, Yang, Jiang (bib0022) 2022 Gregor, LeCun (bib0021) 2010 Mittal, Moorthy, Bovik (bib0035) 2012; 21 Li, Liu, Yang, Sun, Guo (bib0013) 2018; 27 Zhang, Zuo, Gu, Zhang (bib0027) 2017 Land (bib0001) 1983; 80 Xu, Hou, Ren, Liu, Zhu, Yu, Wang, Shao (bib0002) 2020; 29 Xu, Hu, Hu, Jing, Cai, Lu (bib0010) 2025; 158 Martin, Fowlkes, Tal, Malik (bib0028) 2001; 2 Zhao, Chen, Zhang, Wang, Bai (bib0023) 2024; 10 Liu, Wu, Wang (bib0031) 2023; 32 Zhao, Wang, Zhang, Wang, Bai (bib0019) 2023; 524 Fu, Zeng, Huang, Zhang, Ding (bib0005) 2016 Ma, Ma, Liu, Fan, Luo (bib0032) 2022 Mittal, Soundararajan, Bovik (bib0034) 2012; 20 Du, Liu, Zhao, Xu, Li, You (bib0004) 2023; 136 Land (bib0006) 1977; 237 Xu, Deng, Xu (bib0024) 2022; 29 Sreter, Giryes (bib0025) 2018 Zhang, Zuo, Zhang (bib0037) 2018; 27 Yang (10.1016/j.patcog.2025.112371_bib0020) 2021; 30 Zhang (10.1016/j.patcog.2025.112371_bib0027) 2017 Li (10.1016/j.patcog.2025.112371_bib0013) 2018; 27 Mittal (10.1016/j.patcog.2025.112371_bib0034) 2012; 20 Wei (10.1016/j.patcog.2025.112371_bib0008) 2018 Land (10.1016/j.patcog.2025.112371_bib0006) 1977; 237 Guo (10.1016/j.patcog.2025.112371_bib0012) 2016; 26 Sreter (10.1016/j.patcog.2025.112371_bib0025) 2018 Zhang (10.1016/j.patcog.2025.112371_bib0037) 2018; 27 Gregor (10.1016/j.patcog.2025.112371_bib0021) 2010 Zhao (10.1016/j.patcog.2025.112371_bib0023) 2024; 10 Wen (10.1016/j.patcog.2025.112371_bib0011) 2025; 158 Martin (10.1016/j.patcog.2025.112371_bib0028) 2001; 2 Du (10.1016/j.patcog.2025.112371_bib0004) 2023; 136 Agaian (10.1016/j.patcog.2025.112371_bib0036) 2007; 16 Hai (10.1016/j.patcog.2025.112371_bib0029) 2023; 90 Wu (10.1016/j.patcog.2025.112371_bib0022) 2022 Mittal (10.1016/j.patcog.2025.112371_bib0035) 2012; 21 Zhao (10.1016/j.patcog.2025.112371_bib0019) 2023; 524 Ma (10.1016/j.patcog.2025.112371_bib0032) 2022 Xu (10.1016/j.patcog.2025.112371_bib0002) 2020; 29 Fu (10.1016/j.patcog.2025.112371_bib0009) 2023 Liu (10.1016/j.patcog.2025.112371_bib0033) 2023; 45 Xu (10.1016/j.patcog.2025.112371_bib0024) 2022; 29 Kimmel (10.1016/j.patcog.2025.112371_bib0016) 2003; 52 Xiao (10.1016/j.patcog.2025.112371_bib0017) 2024 Xu (10.1016/j.patcog.2025.112371_bib0030) 2024; 295 Fu (10.1016/j.patcog.2025.112371_bib0005) 2016 Zhao (10.1016/j.patcog.2025.112371_bib0018) 2022; 32 Ren (10.1016/j.patcog.2025.112371_bib0003) 2020; 29 Lin (10.1016/j.patcog.2025.112371_bib0014) 2022; 31 Lore (10.1016/j.patcog.2025.112371_bib0007) 2017; 61 Land (10.1016/j.patcog.2025.112371_bib0001) 1983; 80 Xu (10.1016/j.patcog.2025.112371_bib0010) 2025; 158 Liu (10.1016/j.patcog.2025.112371_bib0031) 2023; 32 Du (10.1016/j.patcog.2025.112371_bib0015) 2023; 116 Zhao (10.1016/j.patcog.2025.112371_bib0026) 2023 |
| References_xml | – volume: 61 start-page: 650 year: 2017 end-page: 662 ident: bib0007 article-title: LLNet: a deep autoencoder approach to natural low-light image enhancement publication-title: Pattern Recognit. – volume: 158 year: 2025 ident: bib0010 article-title: UPT-Flow: multi-scale transformer-guided normalizing flow for low-light image enhancement publication-title: Pattern Recognit. – volume: 52 start-page: 7 year: 2003 end-page: 23 ident: bib0016 article-title: A variational framework for retinex publication-title: Int. J. Comput. Vis. – volume: 27 start-page: 2828 year: 2018 end-page: 2841 ident: bib0013 article-title: Structure-revealing low-light image enhancement via robust retinex model publication-title: IEEE Trans. Image Process. – volume: 237 start-page: 108 year: 1977 end-page: 129 ident: bib0006 article-title: The retinex theory of color vision publication-title: Sci. Am. – volume: 158 year: 2025 ident: bib0011 article-title: An illumination-guided dual attention vision transformer for low-light image enhancement publication-title: Pattern Recognit. – start-page: 211 year: 2024 end-page: 218 ident: bib0017 article-title: An image decomposition-based enhancement using a matrix iterative algorithm publication-title: 2024 12th International Conference on Intelligent Control and Information Processing (ICICIP) – volume: 90 year: 2023 ident: bib0029 article-title: R2rnet: Low-light image enhancement via real-low to real-normal network publication-title: J. Vis. Commun. Image Represent – start-page: 22252 year: 2023 end-page: 22261 ident: bib0009 article-title: Learning a simple low-light image enhancer from paired low-light instances publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition – volume: 29 start-page: 1202 year: 2022 end-page: 1206 ident: bib0024 article-title: Revisiting convolutional sparse coding for image denoising: from a multi-scale perspective publication-title: IEEE Signal Process. Lett. – volume: 136 year: 2023 ident: bib0004 article-title: A new image decomposition approach using pixel-wise analysis sparsity model publication-title: Pattern Recognit. – start-page: 2191 year: 2018 end-page: 2195 ident: bib0025 article-title: Learned convolutional sparse coding publication-title: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) – start-page: 399 year: 2010 end-page: 406 ident: bib0021 article-title: Learning fast approximations of sparse coding publication-title: Proceedings of the 27th international conference on international conference on machine learning – year: 2017 ident: bib0027 article-title: Learning deep CNN denoiser prior for image restoration publication-title: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – volume: 45 start-page: 5953 year: 2023 end-page: 5969 ident: bib0033 article-title: Learning with nested scene modeling and cooperative architecture search for low-light vision publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 32 start-page: 1076 year: 2022 end-page: 1088 ident: bib0018 article-title: RetinexDIP: a unified deep framework for low-light image enhancement publication-title: IEEE Trans. Circuits Syst. Video Technol. – volume: 10 year: 2024 ident: bib0023 article-title: RIRO: From retinex-inspired reconstruction optimization model to deep low-light image enhancement unfolding network publication-title: IEEE Trans. Comput. Imaging – volume: 20 start-page: 209 year: 2012 end-page: 212 ident: bib0034 article-title: Making a “completely blind” image quality analyzer publication-title: IEEE Signal Process. Lett. – year: 2018 ident: bib0008 article-title: Deep retinex decomposition for low-light enhancement publication-title: British Machine Vision Conference – volume: 2 start-page: 416 year: 2001 end-page: 423 ident: bib0028 article-title: A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics publication-title: Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001 – volume: 32 start-page: 8486 year: 2023 end-page: 8499 ident: bib0031 article-title: EFINet: restoration for low-light images via enhancement-fusion iterative network publication-title: IEEE Trans. Circuits Syst. Video Technol. – volume: 116 start-page: 1 year: 2023 end-page: 15 ident: bib0015 article-title: Low-light image enhancement and denoising via dual-constrained retinex model publication-title: Appl. Math. Model. – start-page: 5901 year: 2022 end-page: 5910 ident: bib0022 article-title: Uretinex-Net: retinex-based deep unfolding network for low-light image enhancement publication-title: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition – volume: 295 year: 2024 ident: bib0030 article-title: Degraded structure and hue guided auxiliary learning for low-light image enhancement publication-title: Knowl. Based Syst. – volume: 26 start-page: 982 year: 2016 end-page: 993 ident: bib0012 article-title: LIME: low-light image enhancement via illumination map estimation publication-title: IEEE Trans. Image Process. – start-page: 5637 year: 2022 end-page: 5646 ident: bib0032 article-title: Toward fast, flexible, and robust low-light image enhancement publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) – volume: 80 start-page: 5163 year: 1983 end-page: 5169 ident: bib0001 article-title: Recent advances in retinex theory and some implications for cortical computations color vision and the natural image publication-title: Proc. National Acad. Sci. – volume: 16 start-page: 741 year: 2007 end-page: 758 ident: bib0036 article-title: Transform coefficient histogram-based image enhancement algorithms using contrast entropy publication-title: IEEE Trans. Image Process. – volume: 29 start-page: 5022 year: 2020 end-page: 5037 ident: bib0002 article-title: STAR: a structure and texture aware retinex model publication-title: IEEE Trans. Image Process. – volume: 30 start-page: 2072 year: 2021 end-page: 2086 ident: bib0020 article-title: Sparse gradient regularized deep retinex network for robust low-light image enhancement publication-title: IEEE Trans. Image Process. – start-page: 2368 year: 2023 end-page: 2376 ident: bib0026 article-title: Deep convolutional sparse coding networks for interpretable image fusion publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition – volume: 29 start-page: 5862 year: 2020 end-page: 5876 ident: bib0003 article-title: LR3M: robust low-light enhancement via low-rank regularized retinex model publication-title: IEEE Trans. Image Process. – volume: 31 start-page: 4897 year: 2022 end-page: 4908 ident: bib0014 article-title: Low-light enhancement using a plug-and-play retinex model with shrinkage mapping for illumination estimation publication-title: IEEE Trans. Image Process. – volume: 21 start-page: 4695 year: 2012 end-page: 4708 ident: bib0035 article-title: No-reference image quality assessment in the spatial domain publication-title: IEEE Trans. Image Process. – volume: 27 start-page: 4608 year: 2018 end-page: 4622 ident: bib0037 article-title: FFDNet: toward a fast and flexible solution for CNN-based image denoising publication-title: IEEE Trans. Image Process. – volume: 524 start-page: 126 year: 2023 end-page: 141 ident: bib0019 article-title: Learning deep texture-structure decomposition for low-light image restoration and enhancement publication-title: Neurocomputing – start-page: 2782 year: 2016 end-page: 2790 ident: bib0005 article-title: A weighted variational model for simultaneous reflectance and illumination estimation publication-title: Proceedings of the IEEE conference on computer vision and pattern recognition – year: 2018 ident: 10.1016/j.patcog.2025.112371_bib0008 article-title: Deep retinex decomposition for low-light enhancement – volume: 10 year: 2024 ident: 10.1016/j.patcog.2025.112371_bib0023 article-title: RIRO: From retinex-inspired reconstruction optimization model to deep low-light image enhancement unfolding network publication-title: IEEE Trans. Comput. Imaging doi: 10.1109/TCI.2024.3420942 – start-page: 22252 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0009 article-title: Learning a simple low-light image enhancer from paired low-light instances – volume: 158 year: 2025 ident: 10.1016/j.patcog.2025.112371_bib0011 article-title: An illumination-guided dual attention vision transformer for low-light image enhancement publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2024.111033 – volume: 27 start-page: 4608 issue: 9 year: 2018 ident: 10.1016/j.patcog.2025.112371_bib0037 article-title: FFDNet: toward a fast and flexible solution for CNN-based image denoising publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2018.2839891 – volume: 295 year: 2024 ident: 10.1016/j.patcog.2025.112371_bib0030 article-title: Degraded structure and hue guided auxiliary learning for low-light image enhancement publication-title: Knowl. Based Syst. doi: 10.1016/j.knosys.2024.111779 – start-page: 2191 year: 2018 ident: 10.1016/j.patcog.2025.112371_bib0025 article-title: Learned convolutional sparse coding – volume: 32 start-page: 8486 issue: 12 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0031 article-title: EFINet: restoration for low-light images via enhancement-fusion iterative network publication-title: IEEE Trans. Circuits Syst. Video Technol. doi: 10.1109/TCSVT.2022.3195996 – volume: 29 start-page: 1202 year: 2022 ident: 10.1016/j.patcog.2025.112371_bib0024 article-title: Revisiting convolutional sparse coding for image denoising: from a multi-scale perspective publication-title: IEEE Signal Process. Lett. doi: 10.1109/LSP.2022.3175096 – volume: 90 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0029 article-title: R2rnet: Low-light image enhancement via real-low to real-normal network publication-title: J. Vis. Commun. Image Represent doi: 10.1016/j.jvcir.2022.103712 – start-page: 211 year: 2024 ident: 10.1016/j.patcog.2025.112371_bib0017 article-title: An image decomposition-based enhancement using a matrix iterative algorithm – volume: 30 start-page: 2072 year: 2021 ident: 10.1016/j.patcog.2025.112371_bib0020 article-title: Sparse gradient regularized deep retinex network for robust low-light image enhancement publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2021.3050850 – volume: 20 start-page: 209 issue: 3 year: 2012 ident: 10.1016/j.patcog.2025.112371_bib0034 article-title: Making a “completely blind” image quality analyzer publication-title: IEEE Signal Process. Lett. doi: 10.1109/LSP.2012.2227726 – start-page: 2368 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0026 article-title: Deep convolutional sparse coding networks for interpretable image fusion – start-page: 5637 year: 2022 ident: 10.1016/j.patcog.2025.112371_bib0032 article-title: Toward fast, flexible, and robust low-light image enhancement – volume: 32 start-page: 1076 issue: 3 year: 2022 ident: 10.1016/j.patcog.2025.112371_bib0018 article-title: RetinexDIP: a unified deep framework for low-light image enhancement publication-title: IEEE Trans. Circuits Syst. Video Technol. doi: 10.1109/TCSVT.2021.3073371 – volume: 136 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0004 article-title: A new image decomposition approach using pixel-wise analysis sparsity model publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2022.109241 – volume: 61 start-page: 650 year: 2017 ident: 10.1016/j.patcog.2025.112371_bib0007 article-title: LLNet: a deep autoencoder approach to natural low-light image enhancement publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2016.06.008 – volume: 31 start-page: 4897 year: 2022 ident: 10.1016/j.patcog.2025.112371_bib0014 article-title: Low-light enhancement using a plug-and-play retinex model with shrinkage mapping for illumination estimation publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2022.3189805 – volume: 45 start-page: 5953 issue: 5 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0033 article-title: Learning with nested scene modeling and cooperative architecture search for low-light vision publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 158 year: 2025 ident: 10.1016/j.patcog.2025.112371_bib0010 article-title: UPT-Flow: multi-scale transformer-guided normalizing flow for low-light image enhancement publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2024.111076 – volume: 237 start-page: 108 issue: 6 year: 1977 ident: 10.1016/j.patcog.2025.112371_bib0006 article-title: The retinex theory of color vision publication-title: Sci. Am. doi: 10.1038/scientificamerican1277-108 – volume: 524 start-page: 126 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0019 article-title: Learning deep texture-structure decomposition for low-light image restoration and enhancement publication-title: Neurocomputing doi: 10.1016/j.neucom.2022.12.043 – volume: 116 start-page: 1 year: 2023 ident: 10.1016/j.patcog.2025.112371_bib0015 article-title: Low-light image enhancement and denoising via dual-constrained retinex model publication-title: Appl. Math. Model. doi: 10.1016/j.apm.2022.11.022 – start-page: 5901 year: 2022 ident: 10.1016/j.patcog.2025.112371_bib0022 article-title: Uretinex-Net: retinex-based deep unfolding network for low-light image enhancement – volume: 27 start-page: 2828 issue: 6 year: 2018 ident: 10.1016/j.patcog.2025.112371_bib0013 article-title: Structure-revealing low-light image enhancement via robust retinex model publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2018.2810539 – volume: 21 start-page: 4695 issue: 12 year: 2012 ident: 10.1016/j.patcog.2025.112371_bib0035 article-title: No-reference image quality assessment in the spatial domain publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2012.2214050 – start-page: 399 year: 2010 ident: 10.1016/j.patcog.2025.112371_bib0021 article-title: Learning fast approximations of sparse coding – year: 2017 ident: 10.1016/j.patcog.2025.112371_bib0027 article-title: Learning deep CNN denoiser prior for image restoration – volume: 29 start-page: 5022 year: 2020 ident: 10.1016/j.patcog.2025.112371_bib0002 article-title: STAR: a structure and texture aware retinex model publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2020.2974060 – volume: 80 start-page: 5163 issue: 16 year: 1983 ident: 10.1016/j.patcog.2025.112371_bib0001 article-title: Recent advances in retinex theory and some implications for cortical computations color vision and the natural image publication-title: Proc. National Acad. Sci. doi: 10.1073/pnas.80.16.5163 – volume: 26 start-page: 982 issue: 2 year: 2016 ident: 10.1016/j.patcog.2025.112371_bib0012 article-title: LIME: low-light image enhancement via illumination map estimation publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2016.2639450 – volume: 2 start-page: 416 year: 2001 ident: 10.1016/j.patcog.2025.112371_bib0028 article-title: A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics – volume: 29 start-page: 5862 year: 2020 ident: 10.1016/j.patcog.2025.112371_bib0003 article-title: LR3M: robust low-light enhancement via low-rank regularized retinex model publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2020.2984098 – volume: 52 start-page: 7 year: 2003 ident: 10.1016/j.patcog.2025.112371_bib0016 article-title: A variational framework for retinex publication-title: Int. J. Comput. Vis. doi: 10.1023/A:1022314423998 – volume: 16 start-page: 741 issue: 3 year: 2007 ident: 10.1016/j.patcog.2025.112371_bib0036 article-title: Transform coefficient histogram-based image enhancement algorithms using contrast entropy publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2006.888338 – start-page: 2782 year: 2016 ident: 10.1016/j.patcog.2025.112371_bib0005 article-title: A weighted variational model for simultaneous reflectance and illumination estimation |
| SSID | ssj0017142 |
| Score | 2.4954302 |
| Snippet | •We propose a novel image decomposition-based optimization model for low-light image enhancement, by using the total variation and multi-scale convolutional... |
| SourceID | crossref elsevier |
| SourceType | Index Database Publisher |
| StartPage | 112371 |
| SubjectTerms | Deep neural network Image decomposition Low-light image enhancement Retinex theory |
| Title | A novel image enhancement method based on image decomposition and deep neural networks |
| URI | https://dx.doi.org/10.1016/j.patcog.2025.112371 |
| Volume | 172 |
| WOSCitedRecordID | wos001567653900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: ScienceDirect Freedom Collection 2021 issn: 0031-3203 databaseCode: AIEXJ dateStart: 19950101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.sciencedirect.com omitProxy: false ssIdentifier: ssj0017142 providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtZ3NT9swFMCtrnDYZYOxaWwD-cCtCnIcN04uk6qJCdCEOLCpnCLbsUcROBUtFX_-nj_SpgOhcdgliizHifxznp_9PozQQc10TqQUiaEZS1itGfxSGhauXA8pLzWhwqfM_8HPzorxuDzv9b62sTCLG25t8fBQTv8raigD2C509gW4l41CAdwDdLgCdrj-E_jRwDYLfTOY3Dp3HG2vHFdv8g-nRQ_cxFU7I0GoUWvnVh59t7wtodZ6OnCJLgGfDW7is64Se-5zcro4mOh8tDLljyfC771eiqZT5Eua-9lVjDuLuww07zinRMmZpUlGSbYmOcOpO1H2geaWheNUHonlsENwfTiF6aX5DatyOjxcVV_Pgv3X7LT0GWzd0a6r0ErlWqlCK6_QBuXDsuijjdHJ0fh0aUfiKQv54uPXt8GT3sPv8dc8rZx0FI6LLfQmrhTwKBDeRj1t36G37SkcOArlHfRrhD1w7HHiDnAcgGMPHDc21lgDjgE4dsBxAI5b4O_Rz-9HF9-Ok3haRqJA7M6Tmqmy0ITnuRJCUsnTVBkDP55UghkuDHSCzIXKjJaci6xkRJBccFBo-ZCCnv8B9W1j9UeECZHGSNDtZVmAgsmK3NQkVUrCWt6l699FSdtN1TQkRamew7OLeNuXVVTsgsJWwQB59slPL3zTZ_R6NXq_oP787l7voU21mE9md_txdPwBcVVxbg |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+novel+image+enhancement+method+based+on+image+decomposition+and+deep+neural+networks&rft.jtitle=Pattern+recognition&rft.au=Xiao%2C+Yao&rft.au=Xia%2C+Youshen&rft.date=2026-04-01&rft.issn=0031-3203&rft.volume=172&rft.spage=112371&rft_id=info:doi/10.1016%2Fj.patcog.2025.112371&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_patcog_2025_112371 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0031-3203&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0031-3203&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0031-3203&client=summon |