Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (7): 2077-2083.DOI: 10.11772/j.issn.1001-9081.2019122155

• Virtual reality and multimedia computing • Previous Articles     Next Articles

Image super-resolution reconstruction based on deep progressive back-projection attention network

HU Gaopeng1, CHEN Ziliu1, WANG Xiaoming1,2, ZHANG Kaifang1   

  1. 1. School of Computer and Software Engineering, Xihua University, Chengdu Sichuan 610039, China;
    2. Robotics Research Center, Xihua University, Chengdu Sichuan 610039, China
  • Received:2019-12-24 Revised:2020-03-10 Online:2020-07-10 Published:2020-05-13

基于深度渐进式反投影注意力网络的图像超分辨率重建

胡高鹏1, 陈子鎏1, 王晓明1,2, 张开放1   

  1. 1. 西华大学 计算机与软件工程学院, 成都 610039;
    2. 西华大学 机器人研究中心, 成都 610039
  • 通讯作者: 王晓明
  • 作者简介:胡高鹏(1994-),男,河南商丘人,硕士研究生,主要研究方向:模式识别、图像处理;陈子鎏(1996-),男,广东韶关人,硕士研究生,主要研究方向:模式识别、图像处理;王晓明(1977-),男,四川简阳人,副教授,博士,主要研究方向:模式识别、机器学习、图像处理、计算机视觉;张开放(1994-),男,河南周口人,硕士研究生,主要研究方向:模式识别、图像处理。
  • 基金资助:
    西华大学研究生创新基金资助项目(ycjj2019095)。

Abstract: Focused on the problems of Single Image Super-Resolution (SISR) reconstruction methods, such as the loss of high frequency information during the process of image reconstruction, the introduction of noise during the process of upsampling and the difficulty of determining the interdependence relationships between the channels of the feature map, a deep progressive back-projection attention network was proposed. Firstly, a progressive upsampling method was used to gradually scale the Low Resolution (LR) image to a given magnification in order to alleviate problems such as high-frequency information loss caused by upsampling. Then, at each stage of progressive upsampling, iterative back-projection idea was merged to learn mapping relationship between High Resolution (HR) and LR feature maps and reduce the introduced noise in the upsampling process. Finally, the attention mechanism was used to dynamically allocate attention resources to the feature maps generated at different stages of the progressive back-projection network, so that the interdependence relationships between the feature maps were learned by the network model. Experimental results show that the proposed method can increase the Peak Signal-to-Noise Ratio (PSNR) by up to 3.16 dB and the structural similarity by up to 0.218 4.

Key words: super-resolution, progressive upsampling, back-projection network, attention mechanism, deep learning

摘要: 针对多数单帧图像超分辨率(SISR)方法在重建预测图像时存在高频信息丢失和上采样过程中会引入噪声以及特征图各通道之间的相互依赖关系难以确定等问题,提出了深度渐进式反投影注意力网络。首先使用渐进式上采样方法将低分辨率(LR)图像逐步缩放至给定的倍率,缓解上采样过程中造成的高频信息丢失等问题;然后在渐进式上采样的每个阶段融合迭代反投影思想,学习高分辨率(HR)和LR特征图之间的映射关系并减少上采样过程中引入的噪声;最后使用注意力机制为渐进式反投影网络不同阶段产生的特征图动态分配注意力资源,使网络模型学习到各特征图之间的相互依赖关系。实验结果表明,所提出的方法相比主流的超分辨率方法,峰值信噪比(PSNR)最高可增加3.16 dB,结构相似性最高可提升0.218 4。

关键词: 超分辨率, 渐进式上采样, 反投影网络, 注意力机制, 深度学习

CLC Number: