A Super-resolution Method of Remote Sensing Image Using Transformers

Abstract

This paper proposes a Transformers-based super-resolution method for remote sensing images. Firstly, a remote sensing image super-resolution network based on convolutional neural network and Transformer module is constructed; then the training data is used to train the remote sensing image super-resolution network and the optimized network parameters are obtained; finally, the trained remote sensing image super-resolution network is used to super-resolve low-resolution remote sensing images to obtain high-resolution remote sensing images. Experiments are conducted on the public remote sensing dataset (UC Mercedes) and compared with several traditional super-resolution algorithms. The results show that the present algorithm is highly automated and has improved in both accuracy and efficiency.