End-to-End Super-Resolution for Remote-Sensing Images Using an Improved Multi-Scale Residual Network

被引:26
|
作者
Huan, Hai [1 ]
Li, Pengcheng [2 ]
Zou, Nan [2 ]
Wang, Chao [2 ,3 ]
Xie, Yaqin [2 ]
Xie, Yong [4 ]
Xu, Dongdong [1 ]
机构
[1] Nanjing Univ Informat Sci & Technol, Sch Artificial Intelligence, Nanjing 210044, Peoples R China
[2] Nanjing Univ Informat Sci & Technol, Sch Elect & Informat Engn, Nanjing 210044, Peoples R China
[3] Nanjing Univ Informat Sci & Technol, Jiangsu Key Lab Meteorol Observat & Informat Proc, Nanjing 210044, Peoples R China
[4] Nanjing Univ Informat Sci & Technol, Sch Geog Sci, Nanjing 210044, Peoples R China
基金
中国国家自然科学基金;
关键词
remote sensing; super-resolution reconstruction; pyramidal multi-scale residual network; multi-scale dilation residual block; hierarchical feature fusion structure; complementary block;
D O I
10.3390/rs13040666
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Remote-sensing images constitute an important means of obtaining geographic information. Image super-resolution reconstruction techniques are effective methods of improving the spatial resolution of remote-sensing images. Super-resolution reconstruction networks mainly improve the model performance by increasing the network depth. However, blindly increasing the network depth can easily lead to gradient disappearance or gradient explosion, increasing the difficulty of training. This report proposes a new pyramidal multi-scale residual network (PMSRN) that uses hierarchical residual-like connections and dilation convolution to form a multi-scale dilation residual block (MSDRB). The MSDRB enhances the ability to detect context information and fuses hierarchical features through the hierarchical feature fusion structure. Finally, a complementary block of global and local features is added to the reconstruction structure to alleviate the problem that useful original information is ignored. The experimental results showed that, compared with a basic multi-scale residual network, the PMSRN increased the peak signal-to-noise ratio by up to 0.44 dB and the structural similarity to 0.9776.
引用
收藏
页码:1 / 28
页数:25
相关论文
共 50 条
  • [1] Multi-scale Single Image Super-Resolution with Remote-Sensing Application Using Transferred Wide Residual Network
    Farah Deeba
    Yuanchun Zhou
    Fayaz Ali Dharejo
    Yi Du
    Xuezhi Wang
    She Kun
    [J]. Wireless Personal Communications, 2021, 120 : 323 - 342
  • [2] Multi-scale Single Image Super-Resolution with Remote-Sensing Application Using Transferred Wide Residual Network
    Deeba, Farah
    Zhou, Yuanchun
    Dharejo, Fayaz Ali
    Du, Yi
    Wang, Xuezhi
    Kun, She
    [J]. WIRELESS PERSONAL COMMUNICATIONS, 2021, 120 (01) : 323 - 342
  • [3] AN END-TO-END MULTI-SCALE RESIDUAL RECONSTRUCTION NETWORK FOR IMAGE COMPRESSIVE SENSING
    Liu, Renhe
    Li, Sumei
    Hou, Chunping
    [J]. 2019 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2019, : 2070 - 2074
  • [5] DYNAMIC MULTI-SCALE NETWORK FOR REMOTE SENSING IMAGE SUPER-RESOLUTION
    Yao, Ping
    He, Peng
    Cheng, Siyuan
    Fu, Li
    Guo, Zhihao
    Zhao, Jianghong
    [J]. 2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, : 3766 - 3769
  • [6] Remote sensing image reconstruction using an asymmetric multi-scale super-resolution network
    Huan, Hai
    Zou, Nan
    Zhang, Yi
    Xie, Yaqin
    Wang, Chao
    [J]. JOURNAL OF SUPERCOMPUTING, 2022, 78 (17): : 18524 - 18550
  • [7] Remote Sensing Image Super-Resolution using Multi-Scale Convolutional Neural Network
    Qin, Xing
    Gao, Xiaoqi
    Yue, Keqiang
    [J]. 2018 11TH UK-EUROPE-CHINA WORKSHOP ON MILLIMETER WAVES AND TERAHERTZ TECHNOLOGIES (UCMMT2018), VOL 1, 2018,
  • [8] Remote sensing image reconstruction using an asymmetric multi-scale super-resolution network
    Hai Huan
    Nan Zou
    Yi Zhang
    Yaqin Xie
    Chao Wang
    [J]. The Journal of Supercomputing, 2022, 78 : 18524 - 18550
  • [9] ESPC_NASUnet: An End-to-End Super-Resolution Semantic Segmentation Network for Mapping Buildings From Remote Sensing Images
    Xu, Penglei
    Tang, Hong
    Ge, Jiayi
    Feng, Lin
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2021, 14 : 5421 - 5435
  • [10] ESPC_NASUnet: An End-to-End Super-Resolution Semantic Segmentation Network for Mapping Buildings from Remote Sensing Images
    Xu, Penglei
    Tang, Hong
    Ge, Jiayi
    Feng, Lin
    [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14 : 5421 - 5435