Remote Sensing Image Sharpening Method Based on Parallel Deep Learning Network Architecture

Abstract

The present invention provides a remote sensing image sharpening method based on a parallel deep learning network architecture, comprising the following steps: S1, obtaining the spectral element characteristics of the remote sensing panchromatic image during the sharpening process: establishing a multi-level deep convolutional neural network architecture, Obtain the quantitative relationship between the multi-spectral band and the panchromatic band space-time-spectrum in the remote sensing image; S2. Obtain the texture element characteristics of the remote sensing multi-spectral image during the sharpening process: establish a multi-scale deep convolutional neural network architecture to obtain remote sensing images Texture detail features of different ground object types in the panchromatic band; S3. Obtain remote sensing image sharpening products: integrate the spectral element features and texture element features, and establish a deep learning reconstruction network to obtain remote sensing image sharpening products. The invention is applicable to various remote sensing image sensors, and at the same time can improve the spatial resolution and spectral information fidelity of the remote sensing image in the complex types of ground objects, and solves the problem of rapidly and accurately producing sharpened products from the remote sensing image.

Description

technical field [0001] The invention belongs to the technical field of image super-resolution / sharpening processing, and in particular relates to a remote sensing image sharpening method based on a parallel deep learning network architecture. Background technique [0002] Multispectral remote sensing imagery with high spatial resolution plays an important role in land use / land cover classification, object detection and semantic segmentation. However, current remote sensing images with high spatial resolution may not be available to support regional / global land surface observations due to technical limitations and high cost of sensors. Although panchromatic images acquired together with multispectral remote sensing images have higher spatial resolution, they lack rich spectral information. In order to simultaneously use the fine spatial information of panchromatic images and the rich spectral information of multispectral images to serve social production and life, it is nece...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to address the above problems and provide a remote sensing image sharpening method based on a parallel deep learning network architecture, which can be applied to a variety of remote sensing image sensors, and at the same time can improve the spatial resolution and spectral information of remote sensing images in complex ground object types Fidelity, solving the problem of using remote sensing images to quickly and accurately produce sharpened products

Images