基于Swin Transformer的遥感图像超分辨率重建  

Super-resolution Reconstruction of Remote Sensing Image Based on Swin Transformer

作　　者：孔锐[1] 冉友红 KONG Rui;RAN You-Hong(School of Intelligent Systems Science and Engineering,Jinan University,Zhuhai 519070,China)

机构地区：[1]暨南大学智能科学与工程学院,珠海519070

出　　处：《计算机系统应用》2024年第9期85-94,共10页Computer Systems & Applications

摘　　要：由于遥感图像中的物体具有不确定性,同时不同图像之间的特征信息差异较大,导致现有超分辨率方法重建效果差,因此本文提出一种结合Swin Transformer和N-gram模型的NG-MAT模型来实现遥感图像超分辨率.首先,在原始Transformer计算自注意力的分支上并联多注意力模块,用于提取全局特征信息来激活更多像素.其次,将自然语言处理领域的N-gram模型应用到图像处理领域,用三元N-gram模型来加强窗口之间的信息交互.本文提出的方法在所选取的数据集上,峰值信噪比在放大因子为2、3、4时达到了34.68 dB、31.03 dB、28.99 dB,结构相似度在放大因子为2、3、4时达到了0.9266、0.8444、0.7734,实验结果表明,本文提出的方法各个指标都优于其他同类方法.Due to the uncertainty of objects in remote sensing images and significant differences in feature information between different images,existing super-resolution methods yield poor reconstruction results.Therefore,this study proposes an NG-MAT model that combines the Swin Transformer and the N-gram model to achieve super-resolution of remote sensing images.Firstly,multiple attention modules are connected in parallel on the branch of the original Transformer to extract global feature information for activating more pixels.Secondly,the N-gram model from natural language processing is applied to the field of image processing,utilizing a trigram N-gram model to enhance information interaction between windows.The proposed method achieves peak signal-to-noise ratios of 34.68 dB,31.03 dB,and 28.99 dB at amplification factors of 2,3,and 4,respectively,and structural similarity indices of 0.9266,0.8444,and 0.7734 at the same amplification factors on the selected dataset.Experimental results demonstrate that the proposed method outperforms other similar methods in various metrics.

关 键 词：Swin Transformer 超分辨率 N-GRAM 遥感图像 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]