Remote sensing image super-resolution reconstruction method based on unsupervised multi-stage fusion

Abstract

The invention discloses a remote sensing image super-resolution reconstruction method based on unsupervised multi-stage fusion, which mainly solves the problems that the existing remote sensing image super-resolution reconstruction method is poor in robustness and the texture of a super-divided remote sensing image is fuzzy, and comprises the following implementation steps of: constructing a multi-stage fusion network; generating a non-matched training set; sequentially performing training of three stages of content consistency, perceived content consistency and perceived distribution consistency on the network by utilizing a constructed consistency loss function; and performing super-resolution reconstruction on the remote sensing image. According to the method, the multi-level features in the multi-stage fusion network are effectively utilized, the network is trained by using the non-matched real remote sensing image, so that the method has relatively high robustness, and the perception similarity of the bottom layer of the low-resolution remote sensing image is mined while the content on the low-resolution remote sensing image is reserved, so that more real and clearer textures are obtained.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a remote sensing image super-resolution reconstruction method based on an unsupervised multi-stage fusion network in the technical field of image super-resolution reconstruction. The invention can be used to reconstruct low-resolution images in remote sensing images. Background technique [0002] In the field of remote sensing, image super-resolution is to reconstruct high-resolution images from low-resolution image observations. Among them, the image super-resolution reconstruction method based on instance learning assumes that there is a certain mapping between low-resolution images and corresponding high-resolution images. relationship, by learning this mapping relationship and transferring it to the input low-resolution feature space to reconstruct a high-resolution image. Therefore, remote sensing image super-resolution methods can be used for environmental m...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art, and propose a remote sensing image super-resolution reconstruction method based on an unsupervised multi-stage fusion network, which is used to solve the problem of poor robustness and poor practicability of the existing remote sensing image super-resolution technology. problem, and it is also used to solve the problem of unclear image texture in the remote sensing image after super-resolution reconstruction in the prior art

Images