Abstract
Images that are more similar to the original high-resolution images can be generated by deep neural network-based super-resolution methods than the non-learning-based ones, but the huge and sometimes redundant network structure and parameters make them unbearable. To get high-quality super-resolution results in computation resource-limited scenarios, we propose a lightweight skip concatenated residual channel attention network, LCRCA for image super-resolution. Specifically, we design a light but efficient deep residual block (DRB) which can generate more precise residual information by using more convolution layers under the same computation budget. To enhance the feature maps of DRB, an improved channel attention mechanism named statistical channel attention (SCA) is proposed by introducing channel statistics. Besides, compared with the commonly used skip connections, we propose to use skip concatenation (SC) to build information flows for feature maps of different layers. Finally, DRB, SCA, and SC are efficiently used to form the proposed network LCRCA. Experiments on four test sets show that our method can gain up to 3.2 dB and 0.12 dB over the bicubic interpolation and the representative lightweight method FERN, respectively, and can recover image details more accurately than the compared algorithms. Code can be found at https://github.com/pengcm/LCRCA.
Similar content being viewed by others
Notes
The PSNR/SSIM value below each detail-enlarged subgraph is the quantitative result of the corresponding algorithm on the entire test image, not the subgraph
References
Agustsson E, Timofte R (2017) Ntire 2017 challenge on single image super-resolution: Dataset and study. In: IEEE conference on computer vision and pattern recognition workshops (CVPRW). IEEE, Honolulu, pp 1122–1131
Ahn N, Kang B, Sohn KA (2018) Fast accurate, and lightweight super-resolution with cascading residual network. In: Proceedings of the European conference on computer vision (ECCV). Springer International Publishing, Munich, pp 256–272
Ba J, Caruana R (2014) Do deep nets really need to be deep? arXiv:1312.6184
Bevilacqua M, Roumy A, Guillemot C, Alberi-Morel ML (2012) Low-complexity single-image super-resolution based on nonnegative neighbor embedding. In: 23rd british machine vision conference, Guildford, England
Cao Y, Wang C, Song C, Tang Y, Li H (2021) Real-time super-resolution system of 4k-video based on deep learning. arXiv:2107.05307
Chen L, Pan J, Jiang J, Zhang J, Han Z, Bao L (2021a) Multi-stage degradation homogenization for super-resolution of face images with extreme degradations. IEEE Trans Image Process 30:5600–5612. https://doi.org/10.1109/TIP.2021.3086595
Chen W, Yao P, Gai S, Da F (2021b) Multi-scale feature aggregation network for image super-resolution. Appl Intell :1–10
Chen Y, He F, Wu Y, Hou N (2017) A local start search algorithm to compute exact hausdorff distance for arbitrary point sets. Pattern Recogn 67:139–148
Chen Y, Liu L, Phonevilay V, Gu K, Xia R, Xie J, Zhang Q, Yang K (2021c) Image super-resolution reconstruction based on feature map attention mechanism. Appl Intell :1–14
Choi JS, Kim M (2017) A deep convolutional neural network with selection units for super-resolution. In: IEEE conference on computer vision and pattern recognition workshops (CVPRW). IEEE, Honolulu, pp 1150–1156
Dai T, Cai J, Zhang Y, Xia S, Zhang L (2019) Second-order attention network for single image super-resolution. In: IEEE/CVF conference on computer vision and pattern recognition (CVPR). IEEE, Long Beach, pp 11057–11066
Dong C, Loy CC, He K, Tang X (2014) Learning a deep convolutional network for image super-resolution. In: Proceedings of the european conference on computer vision (ECCV). Springer International Publishing, Zurich, pp 184–199
Dong C, Loy CC, Tang X (2016) Accelerating the super-resolution convolutional neural network. In: Proceedings of the European conference on computer vision (ECCV). Springer International Publishing, Amsterdam, pp 391–407
Dong W, Zhou C, Wu F, Wu J, Shi G, Li X (2021) Model-guided deep hyperspectral image super-resolution. IEEE Trans Image Process 30:5754–5768. https://doi.org/10.1109/TIP.2021.3078058
Du X (2021) Single image super-resolution using global enhanced upscale network. Appl Intell :1–7
Fang C, Zhu Y, Liao L, Ling X (2021) Tsrgan: Real-world text image super-resolution based on adversarial learning and triplet attention. Neurocomputing 455:88–96. https://doi.org/10.1016/j.neucom.2021.05.060, https://www.sciencedirect.com/science/article/pii/S0925231221008134
Feng X, Li X, Li J (2021) Multi-scale fractal residual network for image super-resolution. Appl Intell 51(4):1845–1856
Friedman M (1937) The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J Am Stat Assoc 32(200):675–701
Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Proceedings of the fourteenth international conference on artificial intelligence and statistics, JMLR Workshop and Conference Proceedings, Fort Lauderdale, FL, USA, pp 315– 323
Greff K, Srivastava R, Schmidhuber J (2017) Highway and residual networks learn unrolled iterative estimation. arXiv:1612.07771
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: IEEE international conference on computer vision (ICCV). IEEE, Santiago, pp 1026–1034
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: IEEE conference on computer vision and pattern recognition (CVPR). IEEE, Las Vegas, pp 770-778
Hu J, Shen L, Albanie S, Sun G, Wu E (2020a) Squeeze-and-excitation networks. IEEE Trans Pattern Anal Mach Intell 42:2011–2023
Hu X, Fan Z, Jia X, Li Z, Zhang X, Qi L, Xuan Z (2021) Towards effective learning for face super-resolution with shape and pose perturbations. https://doi.org/10.1016/j.knosys.2021.106938. https://www.sciencedirect.com/science/article/pii/S095070512100201X, vol 220, p 106938
Hu Y, Li J, Huang Y, Gao X (2020b) Channel-wise and spatial feature modulation network for single image super-resolution. IEEE Trans Circ Syst Vid Technol 30(11):3911–3927. https://doi.org/10.1109/TCSVT.2019.2915238
Huang JB, Singh A, Ahuja N (2015) Single image super-resolution from transformed self-exemplars. In: IEEE conference on computer vision and pattern recognition (CVPR), Boston, pp 5197-5206
Hui Z, Wang X, Gao X (2018) Fast and accurate single image super-resolution via information distillation network. In: IEEE/CVF conference on computer vision and pattern recognition. IEEE, Salt Lake City, pp 723-731. https://doi.org/10.1109/CVPR.2018.00082
Hui Z, Gao X, Yang Y, Wang X (2019) Lightweight image super-resolution with information multi-distillation network. In: 27th ACM international conference on multimedia (MM), Association for Computing Machinery, Nice, France, MM-19, pp 2024–2032. https://doi.org/10.1145/3343031.3351084
Hui Z, Gao X, Wang X (2020) Lightweight image super-resolution with feature enhancement residual network. Neurocomputing 404:50–60
Isallari M, Rekik I (2021) Brain graph super-resolution using adversarial graph neural network with application to functional brain connectivity, vol 71, p 102084. https://doi.org/10.1016/j.media.2021.102084. https://www.sciencedirect.com/science/article/pii/S1361841521001304
Kim J, Lee JK, Lee KM (2016) Accurate image super-resolution using very deep convolutional networks. In: IEEE conference on computer vision and pattern recognition (CVPR), IEEE, Las Vegas, pp 1646-1654
Kim J, Lee JK, Lee KM (2016) Deeply-recursive convolutional network for image super-resolution. In: IEEE conference on computer vision and pattern recognition (CVPR), IEEE, Las Vegas, pp 1637–1645
Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv:14126980
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097–1105
Kumar V, Gupta P (2012) Importance of statistical measures in digital image processing. Int J Emerg Technol Adv Eng 2(8):56–62
Lai WS, Huang JB, Ahuja N, Yang MH (2017) Deep laplacian pyramid networks for fast and accurate super-resolution
Lee CY, Xie S, Gallagher PW, Zhang Z, Tu Z (2015) Deeply-supervised nets. arXiv:1409.5185
Li B, Liu J, Wang B, Qi Z, Shi Y (2019) S-lwsr: Super lightweight super-resolution network. arXiv:1909.10774
Liang Y, He F, Li H (2019) An asymmetric and optimized encryption method to protect the confidentiality of 3d mesh model. Adv Eng Inform 42:100963
Liang Y, He F, Zeng X (2020) 3d mesh simplification with feature preservation based on whale optimization algorithm and differential evolution. Integr Comput-Aid Eng (Preprint):417– 435
Lim B, Son S, Kim H, Nah S, Lee KM (2017) Enhanced deep residual networks for single image super-resolution. In: IEEE conference on computer vision and pattern recognition workshops (CVPRW), IEEE, Honolulu, pp 1132-1140
Liu B, Ait-Boudaoud D (2020) Effective image super resolution via hierarchical convolutional neural network. Neurocomputing 374:109–116
Lu E, Hu X (2021) Image super-resolution via channel attention and spatial attention. Appl Intell :1–9
Martin DR, Fowlkes CC, Tal D, Malik J (2001) A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In: 8th IEEE international conference on computer vision (ICCV 2001), Vancouver, Canada, vol 2. pp 416-423
Nemenyi PB (1963) Distribution-free multiple comparisons. Princeton University
Paszke A, Gross S, Chintala S, Chanan G, Yang E, Devito Z, Lin Z, Desmaison A, Antiga L, Lerer A (2017) Automatic differentiation in pytorch
Qin J, Huang Y, Wen W (2020) Multi-scale feature fusion residual network for single image super-resolution. Neurocomputing 379:334–342
Shi W, Caballero J, Huszár F, Totz J, Aitken AP, Bishop R, Rueckert D, Wang Z (2016) Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network. In: IEEE conference on computer vision and pattern recognition (CVPR), IEEE, Las Vegas, pp 1874-1883
Park Sung Cheol, Park Min Kyu, Kang Moon Gi (2003) Super-resolution image reconstruction: a technical overview. IEEE Signal Process Mag 20(3):21–36
Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A (2015) Going deeper with convolutions. In: 2015 IEEE conference on computer vision and pattern recognition (CVPR), IEEE, Boston, pp 1-9
Tai Y, Yang J, Liu X, Xu C (2017) Memnet: A persistent memory network for image restoration. In: IEEE international conference on computer vision (ICCV), IEEE, Venice, pp 4549–4557
Timofte R, Smet VD, Gool LV (2014) A+: Adjusted anchored neighborhood regression for fast super-resolution. In: 12th Asian conference on computer vision (ACCV), Singapore
Tong T, Li G, Liu X, Gao Q (2017) Image super-resolution using dense skip connections. In: IEEE international conference on computer vision (ICCV), IEEE, Venice, pp 4809–4817. https://doi.org/10.1109/ICCV.2017.514
Wang C, Li Z, Shi J (2019) Lightweight image super-resolution with adaptive weighted learning network. arXiv:1904.02358
Wang F, Jiang M, Qian C, Yang S, Li C, Zhang H, Wang X, Tang X (2017) Residual attention network for image classification. In: IEEE conference on computer vision and pattern recognition (CVPR), IEEE, Honolulu, pp 6450–6458. https://doi.org/10.1109/CVPR.2017.683
Wang H, Hu Q, Wu C, Chi J, Yu X, Wu H (2021a) Dclnet: Dual closed-loop networks for face super-resolution. Knowl-Based Syst 222:106987. https://doi.org/10.1016/j.knosys.2021.106987, https://www.sciencedirect.com/science/article/pii/S0950705121002501
Wang Q, Gao Q, Wu L, Sun G, Jiao L (2021b) Adversarial multi-path residual network for image super-resolution. IEEE Trans Image Process 30:6648–6658. https://doi.org/10.1109/TIP.2021.3096089
Wenlong Z, Yihao L, Dong C, Qiao Y (2021) Ranksrgan: Generative adversarial networks with ranker for image super-resolution. IEEE Trans Pattern Anal Mach Intell. Early access. https://doi.org/10.1109/TPAMI.2021.3096327
Wu Y, He F, Zhang D, Li X (2015) Service-oriented feature-based data exchange for cloud-based design and manufacturing. IEEE Trans Serv Comput 11(2):341–353
Xiong C, Shi X, Gao Z, Wang G (2021) Attention augmented multi-scale network for single image super-resolution. Appl Intell 51(2):935–951
Xue J, Zhao YQ, Bu Y, Liao W, Chan JCW, Philips W (2021) Spatial-spectral structured sparse low-rank representation for hyperspectral image super-resolution. IEEE Trans Image Process 30:3084–3097. https://doi.org/10.1109/TIP.2021.3058590
Yan Y, Ren W, Hu X, Li K, Shen H, Cao X (2021) Srgat: Single image super-resolution with graph attention network. IEEE Trans Image Process 30:4905–4918. https://doi.org/10.1109/TIP.2021.3077135
Zhang Y, Li K, Li K, Wang L, Zhong B, Fu Y (2018) Image super-resolution using very deep residual channel attention networks. In: Proceedings of the European conference on computer vision (ECCV), Springer International Publishing, Munich, pp 294- 310
Zhang Y, Tian Y, Kong Y, Zhong B, Fu Y (2021) Residual dense network for image restoration. IEEE Trans Pattern Anal Mach Intell 43:2480–2495
Zhao X, Liao Y, Li Y, Zhang T, Zou X (2019) Fc2n: Fully channel-concatenated network for single image super-resolution. arXiv:1907.03221
Zhu L, Zhan S, Zhang H (2019) Stacked u-shape networks with channel-wise attention for image super-resolution. Neurocomputing 345:58–66
Zou Y, Zhang L, Liu C, Wang B, Hu Y, Chen Q (2021) Super-resolution reconstruction of infrared images based on a convolutional neural network with skip connections. Opt Lasers Eng 146:106717
Acknowledgements
This work is supported by National Natural Science Foundation of China (61671126).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interests
The authors declare that they have no conflict of interest.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Peng, C., Shu, P., Huang, X. et al. LCRCA: image super-resolution using lightweight concatenated residual channel attention networks. Appl Intell 52, 10045–10059 (2022). https://doi.org/10.1007/s10489-021-02891-5
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-021-02891-5