In the past few years, image super-resolution has achieved remarkable progress, but due to the texture particularity of remote sensing images, the performance in remote sensing images are not very good. We consider that the texture of the same object in remote sensing image is highly similar, so that relevant textures can be transferred from the high resolution remote sensing image to the low resolution remote sensing image. In this paper, we propose a novel super-resolution network for remote sensing image. Specifically, we take a low resolution image and a reference high resolution image as the inputs, then utilize transformer structure to learn the texture transfer which allows the model to enhance the texture details of the low resolution image, to generate the super-resolution image. Experiments on public datasets show that our method achieves visual improvements.