Cargando…

A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images

Super-resolution (SR) algorithms based on deep learning have dominated in various tasks, including medical imaging, street view surveillance and face recognition. In the remote sensing field, most of the current SR methods utilize the low-resolution (LR) images that directly bicubic downsampled the...

Descripción completa

Detalles Bibliográficos
Autores principales: Yan, Li, Chang, Kun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7959284/
https://www.ncbi.nlm.nih.gov/pubmed/33802432
http://dx.doi.org/10.3390/s21051743
_version_ 1783664937887858688
author Yan, Li
Chang, Kun
author_facet Yan, Li
Chang, Kun
author_sort Yan, Li
collection PubMed
description Super-resolution (SR) algorithms based on deep learning have dominated in various tasks, including medical imaging, street view surveillance and face recognition. In the remote sensing field, most of the current SR methods utilize the low-resolution (LR) images that directly bicubic downsampled the high-resolution (HR) images as not only train set but also test set, thus achieving high PSNR/SSIM scores but showing performance drop in application because the degradation model in remote sensing images is subjected to Gaussian blur with unknown parameters. Inspired by multi-task learning strategy, we propose a multiple-blur-kernel super-resolution framework (MSF), in which a multiple-blur-kernel learning module (MLM) optimizes the parameters of the network transferable and sensitive for SR procedures with different blur kernels. Besides, to simultaneously exploit the prior of the large-scale remote sensing images and recurrent information in a single test image, a class-feature capture module (CCM) and an unsupervised learning module (ULM) are leveraged in our framework. Extensive experiments show that our framework outperforms the current state-of-the-art SR algorithms in remotely sensed imagery SR with unknown Gaussian blur kernel.
format Online
Article
Text
id pubmed-7959284
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-79592842021-03-16 A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images Yan, Li Chang, Kun Sensors (Basel) Article Super-resolution (SR) algorithms based on deep learning have dominated in various tasks, including medical imaging, street view surveillance and face recognition. In the remote sensing field, most of the current SR methods utilize the low-resolution (LR) images that directly bicubic downsampled the high-resolution (HR) images as not only train set but also test set, thus achieving high PSNR/SSIM scores but showing performance drop in application because the degradation model in remote sensing images is subjected to Gaussian blur with unknown parameters. Inspired by multi-task learning strategy, we propose a multiple-blur-kernel super-resolution framework (MSF), in which a multiple-blur-kernel learning module (MLM) optimizes the parameters of the network transferable and sensitive for SR procedures with different blur kernels. Besides, to simultaneously exploit the prior of the large-scale remote sensing images and recurrent information in a single test image, a class-feature capture module (CCM) and an unsupervised learning module (ULM) are leveraged in our framework. Extensive experiments show that our framework outperforms the current state-of-the-art SR algorithms in remotely sensed imagery SR with unknown Gaussian blur kernel. MDPI 2021-03-03 /pmc/articles/PMC7959284/ /pubmed/33802432 http://dx.doi.org/10.3390/s21051743 Text en © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Yan, Li
Chang, Kun
A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images
title A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images
title_full A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images
title_fullStr A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images
title_full_unstemmed A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images
title_short A New Super Resolution Framework Based on Multi-Task Learning for Remote Sensing Images
title_sort new super resolution framework based on multi-task learning for remote sensing images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7959284/
https://www.ncbi.nlm.nih.gov/pubmed/33802432
http://dx.doi.org/10.3390/s21051743
work_keys_str_mv AT yanli anewsuperresolutionframeworkbasedonmultitasklearningforremotesensingimages
AT changkun anewsuperresolutionframeworkbasedonmultitasklearningforremotesensingimages
AT yanli newsuperresolutionframeworkbasedonmultitasklearningforremotesensingimages
AT changkun newsuperresolutionframeworkbasedonmultitasklearningforremotesensingimages