一种轻量级的多尺度通道注意图像超分辨率重建网络  

Image Super-Resolution Reconstruction Based on Lightweight Multi-Scale Channel Attention Network

作　　者：周登文[1] 李文斌 李金新 黄志勇 ZHOU Deng-wen;LI Wen-bin;LI Jin-xin;HUANG Zhi-yong(School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华北电力大学控制与计算机工程学院,北京102206

出　　处：《电子学报》2022年第10期2336-2346,共11页Acta Electronica Sinica

摘　　要：近年来,基于深度卷积神经网络的图像超分辨率技术取得了突出进展,并主导了当前的超分辨率技术的研究.但是,性能的改进,往往以参数量的急剧增加为代价,这限制了超分辨率方法的实际应用.本文设计了一个轻量级单图像超分辨率深度卷积网络,主要贡献包括:提出了一个多尺度的特征融合模块,使用不同感受野的卷积核,提取多种尺度的特征;提出了一个通道搅乱注意力模块,促进特征通道之间的信息流动,并增强特征选择能力;提出了一个全局特征融合连接模块,提高浅层特征的利用率.实验证明,本文方法与当前代表性的方法MSRN(Multi-Scale Residual Network)相比,参数量减少了3/4,重建的高分辨率图像的主观和客观质量均显著更好.Recently,image super-resolution technology based on deep convolutional neural network has made re⁃markable achievements and has become popular in the current super-resolution technology.However,superior performance is often at the expense of the large number of parameter amounts,which limits the real-world applications for single image super-resolution.In this paper,a lightweight single image super-resolution deep convolutional network is proposed.The main contributions of this paper are as follows:a multi-scale feature fusion block is proposed to extract multiple features via convolution kernels with different receptive fields;the channel shuffle attention mechanism we designed promotes the flow of the information across feature channels,which enhances the ability of feature selection;a global feature fusion connec⁃tion is proposed to improve the feature utilization.Extensive experiments demonstrate that the parameter amounts of our method reduced by 3/4 compared with the current state-of-the-art MSRN method,while subjective visual and objective qual⁃ity of the reconstructed high-resolution image are perform significantly better.

关 键 词：超分辨率 深度学习 卷积神经网络 注意力机制 多尺度特征 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]