[1]沈学利,翟宇琦,关刘美,等.多尺度注意力特征融合的单图像超分辨率研究[J].计算机技术与发展,2024,34(07):31-39.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0088]
　SHEN Xue-li,ZHAI Yu-qi,GUAN Liu-mei,et al.Research on Single Image Super-resolution Based on Multi-scale Attention Feature Fusion[J].,2024,34(07):31-39.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0088]
点击复制

多尺度注意力特征融合的单图像超分辨率研究()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
34
期数:
2024年07期
页码:
31-39
栏目:
媒体计算
出版日期:
2024-07-10

文章信息/Info

Title:
Research on Single Image Super-resolution Based on Multi-scale Attention Feature Fusion
文章编号:
1673-629X(2024)07-0031-09
作者:
沈学利; 翟宇琦; 关刘美; 苏婷
辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105
Author(s):
SHEN Xue-li; ZHAI Yu-qi; GUAN Liu-mei; SU Ting
School of Software,Liaoning Technical University,Huludao 125105,China
关键词:
生成对抗网络; 图像超分辨率; 多尺度注意力特征融合; 大核分解; 全局学习与下采样; 令牌
Keywords:
generative adversarial network; image super - resolution; multi - scale attention feature fusion;  large kernel decomposition; global learning and down-sampling; token
分类号:
TP391.4
DOI:
10.20165/j.cnki.ISSN1673-629X.2024.0088
摘要:
高分辨率意味着图像具有高像素密度,可以提供更多的细节,这些细节往往在应用中起到关键作用。 基于生成对抗网络的图像超分辨率由于具有生成丰富细节的潜力,近年来受到越来越多的关注。 针对现有的网络模型忽略从特征中学习本质纹理特征和感受野有限的问题,基于 Real-ESRGAN 和多尺度注意力特征融合,对网络进行优化,将残差稠密块替换成大核分解和多尺度学习相结合模块与全局学习与下采样模块的双分支结构方法,提出一种多尺度注意力融合的单图像超分辨率重建算法,增强每个局部与全局令牌对之间的交互,从而形成更丰富和信息量更大的表示。 对数据集进行2,3,4 倍超分辨率重建实验,通过峰值信噪比(PSNR)、结构相似性(SSIM)对重建结果进行评价,与 SRCNN、SRGAN、ACMF、MSRDN、WYD、LBW、YJX、Real-ESRGAN 等方法进行对比。 结果表明,该算法优于其他模型,且具有更好的直观视觉效果。
Abstract:
High resolution means that the image has a high pixel density,which can provide more details,which often play a key role in the application. Image super-resolution based on generative adversarial networks has attracted more and more attention in recent years due to its potential to generate rich details. Aiming at the problem that the existing network model ignores the learning of essential texture features from features and the limited receptive field,based on Real-ESRGAN and multi-scale attention feature fusion,the network is op-timized,and the residual-in-residual dense block is replaced by a large kernel decomposition and multi-scale learning. The method of combining the module with the dual branch structure of the global learning and down-sampling module proposes a single image super-resolution reconstruction algorithm based on multi-scale attention fusion,which enhances the interaction between each local and global token pair to form a richer and more informative representation. Super-resolution reconstruction experiments of 2,3,4 times were carried out on the data set. The reconstruction results were evaluated by peak signal-to-noise ratio (PSNR) and structural similarity (SSIM),and compared with SRCNN,SRGAN,EDSR,RDN,RCAN,HAN,ENLCA,MAN and Real-ESRGAN methods. The results show that the proposed algorithm is better than other models,and has better visual effect.

相似文献/References:

[1]葛广重,杨敏.基于稀疏表示的单幅图像超分辨率重建[J].计算机技术与发展,2013,(09):113.
　GE Guang-zhong,YANG Min.Single Image Super-resolution Reconstruction Based on Sparse Representation[J].,2013,(07):113.
[2]倪申龙,曾接贤,周世健.整体车牌图像超分辨率重建研究[J].计算机技术与发展,2019,29(04):195.[doi:10. 3969 / j. issn. 1673-629X. 2019. 04. 039]
　NI Shen-long,ZENG Jie-xian,ZHOU Shi-jian.Study on Reconstruction of Overall Image of License Plate[J].,2019,29(07):195.[doi:10. 3969 / j. issn. 1673-629X. 2019. 04. 039]
[3]康嘉钰,苏凡军.基于生成对抗网络的长短兴趣推荐模型[J].计算机技术与发展,2020,30(06):35.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 007]
　KANG Jia-yu,SU Fan-jun.A Long-short-term Interests Recommendation Model Based on Generative Adversarial Networks[J].,2020,30(07):35.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 007]
[4]蒋文杰,罗晓曙*,戴沁璇.一种改进的生成对抗网络的图像上色方法研究[J].计算机技术与发展,2020,30(07):56.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 013]
　JIANG Wen-jie,LUO Xiao-shu*,DAI Qin-xuan.Research on an Improved Method of Generative Adversarial Networks Image Coloring[J].,2020,30(07):56.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 013]
[5]周传华,吴幸运,李 鸣.基于 WGAN 单帧人脸图像超分辨率算法[J].计算机技术与发展,2020,30(09):29.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 006]
　ZHOU Chuan-hua,WU Xing-yun,LI Ming.Single Frame Face Images Super-resolution Algorithm Based on WGAN[J].,2020,30(07):29.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 006]
[6]庄兴旺,丁岳伟.多维度注意力和语义再生的文本生成图像模型[J].计算机技术与发展,2020,30(12):27.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 005]
　ZHUANG Xing-wang,DING Yue-wei.Text-to-image Model by Multidimensional Attention and Semantic Regeneration[J].,2020,30(07):27.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 005]
[7]尹玉婷,肖秦琨.基于深度卷积生成对抗网络的图像生成[J].计算机技术与发展,2021,31(04):86.[doi:10. 3969 / j. issn. 1673-629X. 2021. 04. 015]
　YIN Yu-ting,XIAO Qin-kun.Image Generation Based on Deep Convolution GenerativeAdversarial Networks[J].,2021,31(07):86.[doi:10. 3969 / j. issn. 1673-629X. 2021. 04. 015]
[8]程换新,张志浩,刘文翰,等.基于生成对抗网络的图像识别[J].计算机技术与发展,2021,31(06):175.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 031]
　CHENG Huan-xin,ZHANG Zhi-hao,LIU Wen-han,et al.Image Recognition Based on Generative Adversarial Network[J].,2021,31(07):175.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 031]
[9]苑金辉,乔 艳,费烨琳,等.基于深度迁移学习的心脏 MRI 图像左心室分割[J].计算机技术与发展,2021,31(06):35.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 007]
　YUAN Jin-hui,QIAO Yan,FEI Ye-lin,et al.Left Ventricular Segmentation in Cardiac MRI Images Based onDeep Transfer Learning[J].,2021,31(07):35.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 007]
[10]王田丰,胡谷雨,王 睿,等.基于 AAE 的网络性能异常发现[J].计算机技术与发展,2021,31(07):113.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 019]
　WANG Tian-feng,HU Gu-yu,WANG Rui,et al.AAE-based Anomaly Detection for Network Performance[J].,2021,31(07):113.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 019]

更新日期/Last Update: 2024-07-10