Aiming at the problems of blurred remote sensing images containing artefacts and noise after reconstruction by existing susuper-resolution algorithms,a generative adversarial network based on multiple attention mechanisms is proposed.Firstly,an efficient channel attention is introduced into the residual block of the generator to enhance global correlation and improve the feature extraction capability of the model.Secondly,an iterative attention feature fusion module is used to fuse the input image and the high-level semantic feature map generated by the generator,replacing the common summing operation of long jump connections,reducing the information losscaused by the summing operation of the input image and making the reconstructed image clearer.Finally,the method is validated on different datasets,and the results show that the proposed method improves the peak signal to noise ratio(PSNR)and structural similarity ratio (SSIM)by 0.062～0.122 dB and 0.03～0.08,respectively,compared with the suboptimal algorithm.