Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System
High spatial resolution Earth observation imagery is considered desirable for many scientific and commercial applications. Given repeat multi-angle imagery, an imaging instrument with a specified spatial resolution, we can use image processing and deep learning techniques to enhance the spatial reso...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-12-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | http://www.mdpi.com/2072-4292/11/1/52 |
id |
doaj-dc665ad7883c44c28eda37a5f84fdc1e |
---|---|
record_format |
Article |
spelling |
doaj-dc665ad7883c44c28eda37a5f84fdc1e2020-11-24T21:47:58ZengMDPI AGRemote Sensing2072-42922018-12-011115210.3390/rs11010052rs11010052Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN SystemYu Tao0Jan-Peter Muller1Imaging Group, Mullard Space Science Laboratory, University College London, Holmbury St Mary, Dorking, Surrey RH56NT, UKImaging Group, Mullard Space Science Laboratory, University College London, Holmbury St Mary, Dorking, Surrey RH56NT, UKHigh spatial resolution Earth observation imagery is considered desirable for many scientific and commercial applications. Given repeat multi-angle imagery, an imaging instrument with a specified spatial resolution, we can use image processing and deep learning techniques to enhance the spatial resolution. In this paper, we introduce the University College London (UCL) MAGiGAN super-resolution restoration (SRR) system based on multi-angle feature restoration and deep SRR networks. We explore the application of MAGiGAN SRR to a set of 9 MISR red band images (275 m) to produce up to a factor of 3.75 times resolution enhancement. We show SRR results over four different test sites containing different types of image content including urban and rural targets, sea ice and a cloud field. Different image metrics are introduced to assess the overall SRR performance, and these are employed to compare the SRR results with the original MISR input images and higher resolution Landsat images, where available. Significant resolution improvement over various types of image content is demonstrated and the potential of SRR for different scientific application is discussed.http://www.mdpi.com/2072-4292/11/1/52MISRsuper-resolution restorationSRRfeature matchingGotchaGPTgenerative adversarial networkGANdeep learning |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Yu Tao Jan-Peter Muller |
spellingShingle |
Yu Tao Jan-Peter Muller Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System Remote Sensing MISR super-resolution restoration SRR feature matching Gotcha GPT generative adversarial network GAN deep learning |
author_facet |
Yu Tao Jan-Peter Muller |
author_sort |
Yu Tao |
title |
Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System |
title_short |
Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System |
title_full |
Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System |
title_fullStr |
Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System |
title_full_unstemmed |
Super-Resolution Restoration of MISR Images Using the UCL MAGiGAN System |
title_sort |
super-resolution restoration of misr images using the ucl magigan system |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2018-12-01 |
description |
High spatial resolution Earth observation imagery is considered desirable for many scientific and commercial applications. Given repeat multi-angle imagery, an imaging instrument with a specified spatial resolution, we can use image processing and deep learning techniques to enhance the spatial resolution. In this paper, we introduce the University College London (UCL) MAGiGAN super-resolution restoration (SRR) system based on multi-angle feature restoration and deep SRR networks. We explore the application of MAGiGAN SRR to a set of 9 MISR red band images (275 m) to produce up to a factor of 3.75 times resolution enhancement. We show SRR results over four different test sites containing different types of image content including urban and rural targets, sea ice and a cloud field. Different image metrics are introduced to assess the overall SRR performance, and these are employed to compare the SRR results with the original MISR input images and higher resolution Landsat images, where available. Significant resolution improvement over various types of image content is demonstrated and the potential of SRR for different scientific application is discussed. |
topic |
MISR super-resolution restoration SRR feature matching Gotcha GPT generative adversarial network GAN deep learning |
url |
http://www.mdpi.com/2072-4292/11/1/52 |
work_keys_str_mv |
AT yutao superresolutionrestorationofmisrimagesusingtheuclmagigansystem AT janpetermuller superresolutionrestorationofmisrimagesusingtheuclmagigansystem |
_version_ |
1725894312090140672 |