Lightweight image super-resolution reconstruction network based on Transformer-CNN

doi:10.11772/j.issn.1001-9081.2023010048

Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (1): 292-299.DOI: 10.11772/j.issn.1001-9081.2023010048

• Multimedia computing and computer simulation • Previous Articles

Lightweight image super-resolution reconstruction network based on Transformer-CNN

Hao CHEN¹, Zhenping XIA¹(), Cheng CHENG¹, Xing LIN-LI², Bowen ZHANG¹

^1.School of Electronic and Information Engineering，Suzhou University of Science and Technology，Suzhou Jiangsu 215009，China
^2.School of Physical Science and Technology，Suzhou University of Science and Technology，Suzhou Jiangsu 215009，China

Received:2023-01-17 Revised:2023-04-12 Accepted:2023-04-13 Online:2023-06-06 Published:2024-01-10
Contact: Zhenping XIA
About author:CHEN Hao， born in 2000， M. S. candidate. His research interests include deep learning， digital image processing.
CHENG Cheng， born in 1980， Ph. D.， lecturer. His research interests include machine learning， artificial intelligence.
LIN-LI Xing， born in 1997， M. S. candidate. His research interests include machine vision， deep learning.
ZHANG Bowen， born in 1998， M. S. candidate. His research interests include deep learning， digital image processing
Supported by:
National Natural Science Foundation of China(62002254);Natural Science Foundation of Jiangsu Province(BK20200988);Postgraduate Research and Practice Innovation Program of Jiangsu Province(SJCX21_1424)

基于Transformer-CNN的轻量级图像超分辨率重建网络

陈豪¹, 夏振平¹(), 程成¹, 林李兴², 张博文¹

^1.苏州科技大学电子与信息工程学院，江苏苏州 215009
^2.苏州科技大学物理科学与技术学院，江苏苏州 215009

通讯作者: 夏振平
作者简介:陈豪（2000—），男，四川广安人，硕士研究生，主要研究方向：深度学习、数字图像处理；
程成（1980—），男，江苏苏州人，讲师，博士，主要研究方向：机器学习、人工智能；
林李兴（1997—），男，福建三明人，硕士研究生，主要研究方向：机器视觉、深度学习；
张博文（1998—），男，江苏南通人，硕士研究生，主要研究方向：深度学习、数字图像处理。
第一联系人：夏振平（1985—），男，江苏兴化人，副教授，博士，主要研究方向：图像质量测量、评价及优化；
基金资助:
国家自然科学基金资助项目(62002254);江苏省自然科学基金资助项目(BK20200988);江苏省研究生科研与实践创新计划项目(SJCX21_1424)

Abstract

Abstract:

Aiming at the high computational complexity and large memory consumption of the existing super-resolution reconstruction networks， a lightweight image super-resolution reconstruction network based on Transformer-CNN was proposed， which made the super-resolution reconstruction network more suitable to be applied on embedded terminals such as mobile platforms. Firstly， a hybrid block based on Transformer-CNN was proposed， which enhanced the ability of the network to capture local-global depth features. Then， a modified inverted residual block， with special attention to the characteristics of the high-frequency region， was designed， so that the improvement of feature extraction ability and reduction of inference time were realized. Finally， after exploring the best options for activation function， the GELU （Gaussian Error Linear Unit） activation function was adopted to further improve the network performance. Experimental results show that the proposed network can achieve a good balance between image super-resolution performance and network complexity， and reaches inference speed of 91 frame/s on the benchmark dataset Urban100 with scale factor of 4， which is 11 times faster than the excellent network called SwinIR （Image Restoration using Swin transformer）， indicates that the proposed network can efficiently reconstruct the textures and details of the image and reduce a significant amount of inference time.

Key words: image super-resolution, deep learning, Transformer, Convolutional Neural Network (CNN), lightweight

摘要：

针对现有超分辨率重建网络具有较高的计算复杂度和存在大量内存消耗的问题，提出了一种基于Transformer-CNN的轻量级图像超分辨率重建网络，使超分辨率重建网络更适合应用于移动平台等嵌入式终端。首先，提出了一个基于Transformer-CNN的混合模块，从而增强网络捕获局部-全局深度特征的能力；其次，提出了一个改进的倒置残差块来特别关注高频区域的特征，以提升特征提取能力和减少推理时间；最后，在探索激活函数的最佳选择后，采用GELU （Gaussian Error Linear Unit）激活函数来进一步提高网络性能。实验结果表明，所提网络可以在图像超分辨率性能和网络复杂度之间取得很好的平衡，而且在基准数据集Urban100上4倍超分辨率的推理速度达到91 frame/s，比优秀网络SwinIR （Image Restoration using Swin transformer）快11倍，表明所提网络能够高效地重建图像的纹理和细节，并减少大量的推理时间。

关键词: 图像超分辨率, 深度学习, Transformer, 卷积神经网络, 轻量级

CLC Number:

TP391.41

Hao CHEN, Zhenping XIA, Cheng CHENG, Xing LIN-LI, Bowen ZHANG. Lightweight image super-resolution reconstruction network based on Transformer-CNN[J]. Journal of Computer Applications, 2024, 44(1): 292-299.

陈豪, 夏振平, 程成, 林李兴, 张博文. 基于Transformer-CNN的轻量级图像超分辨率重建网络[J]. 《计算机应用》唯一官方网站, 2024, 44(1): 292-299.

Figures/Tables 12

References 42

1	FANG J， XIAO J， WANG X， et al. Arbitrary scale super resolution network for satellite imagery ［J］. China Communications， 2022， 19（8）： 234-246. 10.23919/jcc.2022.08.017
2	GE L， BAO W， YUAN D， et al. Edge-assisted deep video denoising and super-resolution for real-time surveillance at night ［C］// Proceedings of the 28th Annual International Conference on Mobile Computing and Networking. New York： ACM， 2022： 783-785. 10.1145/3495243.3558754
3	王一宁，赵青杉，秦品乐，等.基于轻量密集神经网络的医学图像超分辨率重建算法［J］.计算机应用， 2022， 42（8）： 2586-2592. 10.11772/j.issn.1001-9081.2021061093
	WANG Y N， ZHAO Q S， QIN P L， et al. Super-resolution reconstruction algorithm of medical image based on lightweight dense neural network ［J］. Journal of Computer Applications， 2022， 42（8）： 2586-2592. 10.11772/j.issn.1001-9081.2021061093
4	陈宗航，胡海龙，姚剑敏，等.基于改进生成对抗网络的单帧图像超分辨率重建［J］.液晶与显示， 2021， 36（5）： 705-712. 10.37188/CJLCD.2020-0250
	CHEN Z H， HU H L， YAO J M， et al. Single frame image super-resolution reconstruction based on improved generative adversarial network ［J］. Chinese Journal of Liquid Crystals and Displays， 2021， 36（5）： 705-712. 10.37188/CJLCD.2020-0250
5	DONG C， LOY C C， HE K， et al. Learning a deep convolutional network for image super-resolution ［C］// Proceedings of the 2014 European Conference on Computer Vision， LNCS 8692. Cham： Springer， 2014： 184-199.
6	HE K， ZHANG X， REN S， et al. Deep residual learning for image recognition ［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 770-778. 10.1109/cvpr.2016.90
7	LEDIG C， THEIS L， HUSZÁR F， et al. Photo-realistic single image super-resolution using a generative adversarial network ［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2017： 105-114. 10.1109/cvpr.2017.19
8	LIM B， SON S， KIM H， et al. Enhanced deep residual networks for single image super-resolution ［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Piscataway： IEEE， 2017： 1132-1140. 10.1109/cvprw.2017.151
9	黄友文，唐欣，周斌.结合双注意力和结构相似度量的图像超分辨率重建网络［J］.液晶与显示， 2022， 37（3）： 367-375. 10.37188/CJLCD.2021-0178
	HUANG Y W， TANG X， ZHOU B. Image super-resolution reconstruction network with dual attention and structural similarity measure ［J］. Chinese Journal of Liquid Crystals and Displays， 2022， 37（3）： 367-375. 10.37188/CJLCD.2021-0178
10	AHN N， KANG B， SOHN K A. Fast， accurate， and lightweight super-resolution with cascading residual network ［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11214. Cham： Springer， 2018： 256-272. 10.48550/arXiv.1803.08664
11	HUI Z， GAO X， YANG Y， et al. Lightweight image super-resolution with information multi-distillation network ［C］// Proceedings of the 27th ACM International Conference on Multimedia. New York： ACM， 2019： 2024-2032. 10.1145/3343031.3351084
12	LIU J， TANG J， WU G. Residual feature distillation network for lightweight image super-resolution ［C］// Proceedings of the 2020 European Conference on Computer Vision， LNCS 12537. Cham： Springer， 2020： 41-55.
13	ZHANG K， DANELLJAN M， LI Y， et al. AIM 2020 challenge on efficient super-resolution： methods and results ［C］// Proceedings of the 2020 European Conference on Computer Vision， LNCS 12537. Cham： Springer， 2020： 5-40.
14	DOSOVITSKIY A， BEYER L， KOLESNIKOV A， et al. An image is worth 16x16 words： Transformers for image recognition at scale ［EB/OL］. （2021-06-03）［2022-10-22］. .
15	CARION N， MASSA F， SYNNAEVE G， et al. End-to-end object detection with transformers ［C］// Proceedings of the 2020 European Conference on Computer Vision LNCS 12346. Cham： Springer， 2020： 213-229. 10.1007/978-3-030-58452-8_13
16	STRUDEL R， GARCIA R， LAPTEV I， et al. Segmenter： Transformer for semantic segmentation ［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 7262-7272. 10.1109/iccv48922.2021.00717
17	LI Y， ZHANG K， CAO J， et al. LocalViT： bringing locality to vision transformers ［EB/OL］. （2021-04-12）［2022-10-22］. . 10.1109/iros55552.2023.10342025
18	LIANG J， CAO J， SUN G， et al. SwinIR： image restoration using Swin Transformer ［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 1833-1844. 10.1109/iccvw54120.2021.00210
19	LIU Z， LIN Y， CAO Y， et al. Swin Transformer： hierarchical vision transformer using shifted windows ［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 9992-10002. 10.1109/iccv48922.2021.00986
20	HENDRYCKS D， GIMPEL K. Gaussian error linear units （GELUs）［EB/OL］. （2020-07-08）［2022-10-22］. .
21	ZAMIR S W， ARORA A， KHAN S， et al. Restormer： efficient transformer for high-resolution image restoration ［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2022： 5718-5729. 10.1109/cvpr52688.2022.00564
22	SANDLER M， HOWARD A， ZHU M， et al. MobileNetV2： inverted residuals and linear bottlenecks ［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 4510-4520. 10.1109/cvpr.2018.00474
23	MAO X， LI Q， XIE H， et al. Least squares generative adversarial networks ［C］// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway： IEEE， 2017： 2813-2821. 10.1109/iccv.2017.304
24	WANG X， YU K， WU S， et al. ESRGAN： enhanced super-resolution generative adversarial networks ［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11133. Cham： Springer， 2019： 63-79.
25	WANG X， XIE L， DONG C， et al. Real-ESRGAN： training real-world blind super-resolution with pure synthetic data ［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision Workshops. Piscataway： IEEE， 2021： 1905-1914. 10.1109/iccvw54120.2021.00217
26	DU Z， LIU D， LIU J， et al. Fast and memory-efficient network towards efficient image super-resolution ［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway： IEEE， 2022： 853-862. 10.1109/cvprw56347.2022.00101
27	ZHAO H， KONG X， HE J， et al. Efficient image super-resolution using pixel attention ［C］// Proceedings of the 2021 European Conference on Computer Vision， LNCS 12537. Cham： Springer， 2020： 56-72.
28	TIMOFTE R， AGUSTSSON E， VAN GOOL L， et al. NTIRE 2017 challenge on single image super-resolution： methods and results ［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Piscataway： IEEE， 2017： 1110-1121. 10.1109/cvprw.2017.150
29	BEVILACQUA M， ROUMY A， GUILLEMOT C， et al. Low-complexity single-image super-resolution based on nonnegative neighbor embedding ［C］// Proceedings of the 2012 British Machine Vision Conference. Durham： BMVA Press， 2012： No.135. 10.5244/c.26.135
30	ZEYDE R， ELAD M， PROTTER M. On single image scale-up using sparse-representations ［C］// Proceedings of the 2010 International Conference on Curves and Surfaces LNCS 6920. Cham： Springer， 2012： 711-730. 10.1007/978-3-642-27413-8_47
31	ARBELÁEZ P， MAIRE M， FOWLKES C， et al. Contour detection and hierarchical image segmentation ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2011， 33（5）： 898-916. 10.1109/tpami.2010.161
32	HUANG J B， SINGH A， AHUJA N. Single image super-resolution from transformed self-exemplars ［C］// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2015： 5197-5206. 10.1109/cvpr.2015.7299156
33	MATSUI Y， ITO K， ARAMAKI Y， et al. Sketch-based manga retrieval using Manga109 dataset ［J］. Multimedia Tools and Applications， 2017， 76（20）： 21811-21838. 10.1007/s11042-016-4020-z
34	GAO X， LU W， TAO D， et al. Image quality assessment based on multiscale geometric analysis ［J］. IEEE Transactions on Image Processing， 2009， 18（7）： 1409-1423. 10.1109/tip.2009.2018014
35	WANG Z， BOVIK A C， SHEIKH H R， et al. Image quality assessment： from error visibility to structural similarity ［J］. IEEE Transactions on Image Processing， 2004， 13（4）： 600-612. 10.1109/tip.2003.819861
36	KINGMA D P， BA J L. Adam： a method for stochastic optimization ［EB/OL］. （2017-01-30）［2022-08-03］. .
37	HOWARD A， SANDLER M， CHEN B， et al. Searching for MobileNetV3 ［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 1314-1324. 10.1109/iccv.2019.00140
38	DONG C， LOY C C， TANG X. Accelerating the super-resolution convolutional neural network ［C］// Proceedings of the 2016 European Conference on Computer Vision， LNCS 9906. Cham： Springer， 2016： 391-407.
39	KIM J， LEE J K， LEE K M. Deeply-recursive convolutional network for image super-resolution ［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 1637-1645. 10.1109/cvpr.2016.181
40	LAI W S， HUANG J B， AHUJA N， et al. Fast and accurate image super-resolution with deep Laplacian pyramid networks ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2019， 41（11）： 2599-2613. 10.1109/tpami.2018.2865304
41	ZHANG K， ZUO W， ZHANG L. Learning a single convolutional super-resolution network for multiple degradations ［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 3262-3271. 10.1109/cvpr.2018.00344
42	ZHANG Y， LI K， LI K， et al. Image super-resolution using very deep residual channel attention networks ［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11211. Cham： Springer， 2018： 294-310.

深度	宽度	参数量/10³	乘加量/10⁹	PSNR/dB	SSIM
2	32	109	14.02	30.14	0.830 8
2	40	161	20.84	30.18	0.831 6
4	32	194	24.91	30.31	0.834 8
4	40	290	37.39	30.40	0.8366

深度	宽度	参数量/10³	乘加量/10⁹	PSNR/dB	SSIM
2	32	109	14.02	30.14	0.830 8
2	40	161	20.84	30.18	0.831 6
4	32	194	24.91	30.31	0.834 8
4	40	290	37.39	30.40	0.8366

模块	参数量/10³	乘加量/10⁹	PSNR/dB	SSIM
LTB	240	30.95	31.98	0.892 6
LTB-IRB	293	37.68	32.17	0.894 9
LTB-MIRB	290	37.39	32.29	0.8954

模块	参数量/10³	乘加量/10⁹	PSNR/dB	SSIM
LTB	240	30.95	31.98	0.892 6
LTB-IRB	293	37.68	32.17	0.894 9
LTB-MIRB	290	37.39	32.29	0.8954

激活函数	Set5		Set14		BSD100		Urban100		Manga109
激活函数	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
ReLU	32.19	0.894 1	28.56	0.780 6	27.55	0.735 1	26.01	0.783 7	30.37	0.907 2
LeakyReLU	32.28	0.895 1	28.55	0.780 8	27.54	0.735 4	26.06	0.783 5	30.43	0.907 9
h-swish	32.26	0.895 0	28.58	0.7816	27.55	0.735 6	26.07	0.784 2	30.38	0.907 1
GELU	32.29	0.8954	28.59	0.781 3	27.56	0.7360	26.09	0.7845	30.46	0.9081

Lightweight image super-resolution reconstruction network based on Transformer-CNN

基于Transformer-CNN的轻量级图像超分辨率重建网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 42

Related Articles 15

Recommended Articles

Metrics

算法	参数量/10³	Set5		Set14		BSD100		Urban100		Manga109
算法	参数量/10³	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
Bicubic	—	33.66	0.929 9	30.24	0.868 8	29.56	0.843 1	26.88	0.840 3	30.80	0.933 9
SRCNN^［5］	8	36.66	0.954 2	32.45	0.906 7	31.36	0.887 9	29.50	0.894 6	35.60	0.966 3
FSRCNN^［38］	13	37.00	0.955 8	32.63	0.908 8	31.53	0.892 0	29.88	0.902 0	36.67	0.971 0
DRCN^［39］	1 774	37.63	0.958 8	33.04	0.911 8	31.85	0.894 2	30.75	0.913 3	37.55	0.973 2
LapSRN^［40］	251	37.52	0.959 1	32.99	0.912 4	31.80	0.895 2	30.41	0.910 3	37.27	0.974 0
SRGAN^［7］	1 370	38.05	0.9607	33.64	0.917 8	32.22	0.9002	32.23	0.9295	38.05	0.960 7
SRMDNF^［41］	1 511	37.79	0.960 1	33.32	0.915 9	32.05	0.898 5	31.33	0.920 4	38.07	0.976 1
CARN^［10］	1 592	37.76	0.959 0	33.52	0.916 6	32.09	0.897 8	31.92	0.925 6	38.36	0.976 5
IMDN^［11］	694	38.00	0.960 5	33.63	0.917 7	32.19	0.899 6	32.17	0.928 3	38.88	0.977 4
PAN^［27］	261	38.00	0.960 5	33.59	0.918 1	32.18	0.899 7	32.01	0.927 3	38.70	0.977 3
RFDN^［12］	534	38.05	0.960 6	33.68	0.9184	32.16	0.899 4	32.12	0.927 8	38.88	0.977 3
SwinIR^［18］	878	38.14	0.9611	33.86	0.9206	32.31	0.9012	32.76	0.9340	39.12	0.9783
LHNTC	277	38.04	0.960 6	33.67	0.9184	32.19	0.899 8	32.12	0.927 5	38.88	0.9775

算法	PSNR/dB	SSIM	参数量/10³	平均运行时间/s
RCAN^［42］	26.82	0.808 7	15592	0.260 2
IMDN^［11］	26.04	0.783 8	715	0.021 7
PAN^［27］	26.11	0.785 4	272	0.021 8
RFDN^［12］	26.11	0.785 8	550	0.025 1
SwinIR^［18］	26.47	0.798 0	897	0.134 9
LHNTC	26.09	0.784 5	290	0.0110

[1]	Yunyun GAO, Lasheng ZHAO, Qiang ZHANG. Acoustic word embedding model based on Bi-LSTM and convolutional-Transformer [J]. Journal of Computer Applications, 2024, 44(1): 123-128.
[2]	Junhong ZHU, Junyu LAI, Lianqiang GAN, Zhiyong CHEN, Huashuo LIU, Guoyao XU. Video prediction model combining involution and convolution operators [J]. Journal of Computer Applications, 2024, 44(1): 113-122.
[3]	Yuning ZHANG, Abudukelimu ABULIZI, Tisheng MEI, Chun XU, Maierdana MAIMAITIREYIMU, Halidanmu ABUDUKELIMU, Yutao HOU. Anomaly detection method for skeletal X-ray images based on self-supervised feature extraction [J]. Journal of Computer Applications, 2024, 44(1): 175-181.
[4]	Li’an CHEN, Yi GUO. Text sentiment analysis model based on individual bias information [J]. Journal of Computer Applications, 2024, 44(1): 145-151.
[5]	Hong WANG, Qing QIAN, Huan WANG, Yong LONG. Lightweight image tamper localization algorithm based on large kernel attention convolution [J]. Journal of Computer Applications, 2023, 43(9): 2692-2699.
[6]	Mengmeng CHEN, Zhiwei QIAO. Sparse reconstruction of CT images based on Uformer with fused channel attention [J]. Journal of Computer Applications, 2023, 43(9): 2948-2954.
[7]	Ziyi HE, Yan YANG, Yiling ZHANG. Multi-view clustering network with deep fusion [J]. Journal of Computer Applications, 2023, 43(9): 2651-2656.
[8]	Kunting LU, Rongrong FEI, Xuande ZHANG. Remote sensing image pansharpening by convolutional neural network [J]. Journal of Computer Applications, 2023, 43(9): 2963-2969.
[9]	Guangyao ZHAO, Xuan SHEN, Bo YU, Chenhui YI, Zhen LI. Impossible differential cryptanalysis of reduced-round ultra-lightweight block cipher PFP [J]. Journal of Computer Applications, 2023, 43(9): 2784-2788.
[10]	Shaofa SHANG, Lin JIANG, Yuancheng LI, Yun ZHU. Adaptive partitioning and scheduling method of convolutional neural network inference model on heterogeneous platforms [J]. Journal of Computer Applications, 2023, 43(9): 2828-2835.
[11]	Juntao CHEN, Ziqi ZHU. Image copy-move forgery detection based on multi-scale feature extraction and fusion [J]. Journal of Computer Applications, 2023, 43(9): 2919-2924.
[12]	Xiaolin LI, Songjia YANG. Hybrid beamforming for multi-user mmWave relay networks using deep learning [J]. Journal of Computer Applications, 2023, 43(8): 2511-2516.
[13]	Shengwei DUAN, Xinyu CHENG, Haozhou WANG, Fei WANG. Dam surface disease detection algorithm based on improved YOLOv5 [J]. Journal of Computer Applications, 2023, 43(8): 2619-2629.
[14]	Xiang GUO, Wengang JIANG, Yuhang WANG. Encrypted traffic classification method based on improved Inception-ResNet [J]. Journal of Computer Applications, 2023, 43(8): 2471-2476.
[15]	Yumeng CUI, Jingya WANG, Xiaowen LIU, Shangyi YAN, Zhizhong TAO. General text classification model combining attention and cropping mechanism [J]. Journal of Computer Applications, 2023, 43(8): 2396-2405.

算法	参数量/10³	Set5		Set14		BSD100		Urban100		Manga109
算法	参数量/10³	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
Bicubic	—	30.39	0.868 2	27.55	0.774 2	27.21	0.738 5	24.46	0.734 9	26.95	0.855 6
SRCNN^［5］	8	32.75	0.909 0	29.30	0.821 5	28.41	0.786 3	26.24	0.798 9	30.48	0.911 7
FSRCNN^［38］	13	33.18	0.914 0	29.37	0.824 0	28.53	0.791 0	26.43	0.808 0	31.10	0.921 0
DRCN^［39］	1 774	33.82	0.922 6	29.76	0.831 1	28.80	0.796 3	27.15	0.827 6	32.24	0.934 3
LapSRN^［40］	502	33.81	0.922 0	29.79	0.832 5	28.82	0.798 0	27.07	0.827 5	32.21	0.935 0
SRGAN^［7］	1 554	34.41	0.9274	30.36	0.8427	29.11	0.8055	28.20	0.8535	33.54	0.944 8
SRMDNF^［41］	1 528	34.12	0.925 4	30.04	0.838 2	28.97	0.802 5	27.57	0.839 8	33.00	0.940 3
CARN^［10］	1 592	34.29	0.925 5	30.29	0.840 7	29.06	0.803 4	28.06	0.849 3	33.61	0.944 5
IMDN^［11］	703	34.36	0.927 0	30.32	0.841 7	29.09	0.804 6	28.17	0.851 9	33.61	0.944 5
PAN^［27］	261	34.40	0.927 1	30.36	0.842 3	29.11	0.805 0	28.11	0.851 1	33.61	0.944 8
RFDN^［12］	541	34.41	0.927 3	30.34	0.842 0	29.09	0.805 0	28.21	0.852 5	33.67	0.944 9
SwinIR^［18］	886	34.62	0.9289	30.54	0.8463	29.20	0.8082	28.66	0.8624	33.98	0.9478
LHNTC	283	34.43	0.927 3	30.35	0.842 1	29.06	0.804 5	28.18	0.852 2	33.73	0.9451

算法	参数量/10³	Set5		Set14		BSD100		Urban100		Manga109
算法	参数量/10³	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
Bicubic	—	30.39	0.868 2	27.55	0.774 2	27.21	0.738 5	24.46	0.734 9	26.95	0.855 6
SRCNN^［5］	8	32.75	0.909 0	29.30	0.821 5	28.41	0.786 3	26.24	0.798 9	30.48	0.911 7
FSRCNN^［38］	13	33.18	0.914 0	29.37	0.824 0	28.53	0.791 0	26.43	0.808 0	31.10	0.921 0
DRCN^［39］	1 774	33.82	0.922 6	29.76	0.831 1	28.80	0.796 3	27.15	0.827 6	32.24	0.934 3
LapSRN^［40］	502	33.81	0.922 0	29.79	0.832 5	28.82	0.798 0	27.07	0.827 5	32.21	0.935 0
SRGAN^［7］	1 554	34.41	0.9274	30.36	0.8427	29.11	0.8055	28.20	0.8535	33.54	0.944 8
SRMDNF^［41］	1 528	34.12	0.925 4	30.04	0.838 2	28.97	0.802 5	27.57	0.839 8	33.00	0.940 3
CARN^［10］	1 592	34.29	0.925 5	30.29	0.840 7	29.06	0.803 4	28.06	0.849 3	33.61	0.944 5
IMDN^［11］	703	34.36	0.927 0	30.32	0.841 7	29.09	0.804 6	28.17	0.851 9	33.61	0.944 5
PAN^［27］	261	34.40	0.927 1	30.36	0.842 3	29.11	0.805 0	28.11	0.851 1	33.61	0.944 8
RFDN^［12］	541	34.41	0.927 3	30.34	0.842 0	29.09	0.805 0	28.21	0.852 5	33.67	0.944 9
SwinIR^［18］	886	34.62	0.9289	30.54	0.8463	29.20	0.8082	28.66	0.8624	33.98	0.9478
LHNTC	283	34.43	0.927 3	30.35	0.842 1	29.06	0.804 5	28.18	0.852 2	33.73	0.9451

算法	参数量/10³	Set5		Set14		BSD100		Urban100		Manga109
算法	参数量/10³	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
Bicubic	—	28.42	0.810 4	26.00	0.702 7	25.96	0.667 5	23.14	0.657 7	24.89	0.786 6
SRCNN^［5］	8	30.48	0.862 6	27.50	0.751 3	26.90	0.710 1	24.52	0.722 1	27.58	0.855 5
FSRCNN^［38］	13	30.72	0.866 0	27.61	0.755 0	26.98	0.715 0	24.62	0.728 0	27.90	0.861 0
DRCN^［39］	1 774	31.53	0.885 4	28.02	0.767 0	27.23	0.723 3	25.14	0.751 0	28.93	0.885 4
LapSRN^［40］	502	31.54	0.885 2	28.09	0.770 0	27.32	0.727 5	25.21	0.756 2	29.09	0.890 0
SRGAN^［7］	1 518	32.17	0.895 1	28.61	0.7823	27.59	0.7365	26.12	0.7871	30.48	0.908 7
SRMDNF^［41］	1 552	31.96	0.892 5	28.35	0.778 7	27.49	0.733 7	25.68	0.773 1	30.09	0.902 4
CARN^［10］	1 592	32.13	0.893 7	28.60	0.780 6	27.58	0.734 9	26.07	0.783 7	30.47	0.908 4
IMDN^［11］	715	32.21	0.894 8	28.58	0.781 1	27.56	0.735 3	26.04	0.783 8	30.45	0.907 5
PAN^［27］	272	32.13	0.894 8	28.61	0.782 2	27.59	0.736 3	26.11	0.785 4	30.51	0.909 5
RFDN^［12］	550	32.24	0.895 2	28.61	0.781 9	27.57	0.736 0	26.11	0.785 8	30.58	0.9089
SwinIR^［18］	897	32.44	0.8976	28.77	0.7858	27.69	0.7406	26.47	0.7980	30.92	0.9151
LHNTC	290	32.29	0.8954	28.59	0.781 3	27.56	0.736 0	26.09	0.784 5	30.46	0.908 1

算法	参数量/10³	Set5		Set14		BSD100		Urban100		Manga109
算法	参数量/10³	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM	PSNR/dB	SSIM
Bicubic	—	28.42	0.810 4	26.00	0.702 7	25.96	0.667 5	23.14	0.657 7	24.89	0.786 6
SRCNN^［5］	8	30.48	0.862 6	27.50	0.751 3	26.90	0.710 1	24.52	0.722 1	27.58	0.855 5
FSRCNN^［38］	13	30.72	0.866 0	27.61	0.755 0	26.98	0.715 0	24.62	0.728 0	27.90	0.861 0
DRCN^［39］	1 774	31.53	0.885 4	28.02	0.767 0	27.23	0.723 3	25.14	0.751 0	28.93	0.885 4
LapSRN^［40］	502	31.54	0.885 2	28.09	0.770 0	27.32	0.727 5	25.21	0.756 2	29.09	0.890 0
SRGAN^［7］	1 518	32.17	0.895 1	28.61	0.7823	27.59	0.7365	26.12	0.7871	30.48	0.908 7
SRMDNF^［41］	1 552	31.96	0.892 5	28.35	0.778 7	27.49	0.733 7	25.68	0.773 1	30.09	0.902 4
CARN^［10］	1 592	32.13	0.893 7	28.60	0.780 6	27.58	0.734 9	26.07	0.783 7	30.47	0.908 4
IMDN^［11］	715	32.21	0.894 8	28.58	0.781 1	27.56	0.735 3	26.04	0.783 8	30.45	0.907 5
PAN^［27］	272	32.13	0.894 8	28.61	0.782 2	27.59	0.736 3	26.11	0.785 4	30.51	0.909 5
RFDN^［12］	550	32.24	0.895 2	28.61	0.781 9	27.57	0.736 0	26.11	0.785 8	30.58	0.9089
SwinIR^［18］	897	32.44	0.8976	28.77	0.7858	27.69	0.7406	26.47	0.7980	30.92	0.9151
LHNTC	290	32.29	0.8954	28.59	0.781 3	27.56	0.736 0	26.09	0.784 5	30.46	0.908 1