联合图像—频率监督的图像超分辨率重建算法  

Image super-resolution reconstruction based on joint image-frequency supervision

作　　者：陈书理[1,2,3] 张书贵 赵展[1,2,3] Chen Shuli;Zhang Shugui;Zhao Zhan(School of Information Engineering,Kaifeng University,Kaifeng Henan 475001,China;Research Center of High-Standard Farmland Intelligent Irrigation Project in Henan,Kaifeng Henan 475000,China;Kaifeng Agricultural Internet of Things Engineering Technology Center,Kaifeng Henan 475000,China)

机构地区：[1]开封大学信息工程学院,河南开封475001 [2]河南省高标准农田智能灌溉工程研究中心,河南开封475000 [3]开封市农业物联网工程技术中心,河南开封475000

出　　处：《计算机应用研究》2023年第2期623-627,共5页Application Research of Computers

基　　金：河南省教育厅重点科研项目(22A520001);国家自然科学基金资助项目(61702185);河南省高等学校青年骨干教师培养计划资助项目(2020GZGG043);河南省高等学校重点科研计划项目(21A520029,21A520028)。

摘　　要：针对现有单幅图像超分重建方法难以捕获图像中完备有效信息的问题,提出一种联合图像—频率监督的图像超分辨率重建算法,旨在利用多个域之间的互补信息,进而获得更完备的图像特征表示。首先通过分析图像像素转换到频域空间后的特性,根据其复数表征方式提出了一种新的频域距离监督损失,将频谱信息有效地应用到卷积神经网络的优化过程;然后通过分析频域中不同频带的表征特点,在频域距离损失基础上构建了频谱加权损失,并将其分别应用到低频和高频两个频带;最后结合图像域的监督,构成多个域的联合优化,取得良好的性能。在Set14、B100和Kodak三种公开数据集上进行了验证,结果表明:该算法的PSNR和SSIM分别达到了33.47 dB和0.9859,与几种图像超分方法相比取得了最好的性能。Aiming at the problem that the existing single image super segmentation reconstruction methods are difficult to capture complete and effective information in the image,this paper proposed a joint image-frequency supervised image super-resolution reconstruction algorithm,which aimed to utilize the complementary information between multiple domains,thereby obtaining a more complete image feature representation.Firstly,by analyzing the characteristics of image pixels after converting to frequency domain space,this paper proposed a new frequency domain distance supervision loss according to its complex re-presentation,which effectively applied spectral information to the optimization process of convolutional neural networks.Based on the characterization characteristics of different frequency bands in the domain,this paper constructed the spectral weighting loss on the basis of the distance loss in the frequency domain,and applied it to the two frequency bands of low frequency and high frequency respectively.Finally,combined with the supervision of the image domain,it formed a joint optimization of multiple domains and got good performance.The results on Set14,B100 and Kodak show that the PSNR and SSIM of the proposed algorithm reach 33.47 dB and 0.9859,respectively,and the proposed algorithm achieves the best performance compared with several image super-resolution methods.

关 键 词：单幅图像超分辨率重建 图像—频率联合监督 频域距离监督损失 频谱加权损失 卷积神经网络 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]