DeepEmSat: Deep Emulation for Satellite Data Mining
The growing volume of Earth science data available from climate simulations and satellite remote sensing offers unprecedented opportunity for scientific insight, while also presenting computational challenges. One potential area of impact is atmospheric correction, where physics-based numerical mode...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2019-12-01
|
Series: | Frontiers in Big Data |
Subjects: | |
Online Access: | https://www.frontiersin.org/article/10.3389/fdata.2019.00042/full |
id |
doaj-d608b0df4bdd4f3fb3d4e35324890bd2 |
---|---|
record_format |
Article |
spelling |
doaj-d608b0df4bdd4f3fb3d4e35324890bd22020-11-25T02:28:24ZengFrontiers Media S.A.Frontiers in Big Data2624-909X2019-12-01210.3389/fdata.2019.00042473063DeepEmSat: Deep Emulation for Satellite Data MiningKate Duffy0Kate Duffy1Thomas Vandal2Thomas Vandal3Shuang Li4Shuang Li5Sangram Ganguly6Sangram Ganguly7Ramakrishna Nemani8Auroop R. Ganguly9Sustainability and Data Sciences Laboratory, Department of Civil and Environmental Engineering, Northeastern University, Boston, MA, United StatesAmes Research Center, NASA, Mountain View, CA, United StatesAmes Research Center, NASA, Mountain View, CA, United StatesBay Area Environmental Research Institute, Petaluma, CA, United StatesAmes Research Center, NASA, Mountain View, CA, United StatesBay Area Environmental Research Institute, Petaluma, CA, United StatesAmes Research Center, NASA, Mountain View, CA, United StatesBay Area Environmental Research Institute, Petaluma, CA, United StatesAmes Research Center, NASA, Mountain View, CA, United StatesSustainability and Data Sciences Laboratory, Department of Civil and Environmental Engineering, Northeastern University, Boston, MA, United StatesThe growing volume of Earth science data available from climate simulations and satellite remote sensing offers unprecedented opportunity for scientific insight, while also presenting computational challenges. One potential area of impact is atmospheric correction, where physics-based numerical models retrieve surface reflectance information from top of atmosphere observations, and are too computationally intensive to be run in real time. Machine learning methods have demonstrated potential as fast statistical models for expensive simulations and for extracting credible insights from complex datasets. Here, we develop DeepEmSat: a deep learning emulator approach for atmospheric correction, and offer comparison against physics-based models to support the hypothesis that deep learning can make a contribution to the efficient processing of satellite images.https://www.frontiersin.org/article/10.3389/fdata.2019.00042/fullremote sensingmachine learningdeep learningatmospheric correctionemulator |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Kate Duffy Kate Duffy Thomas Vandal Thomas Vandal Shuang Li Shuang Li Sangram Ganguly Sangram Ganguly Ramakrishna Nemani Auroop R. Ganguly |
spellingShingle |
Kate Duffy Kate Duffy Thomas Vandal Thomas Vandal Shuang Li Shuang Li Sangram Ganguly Sangram Ganguly Ramakrishna Nemani Auroop R. Ganguly DeepEmSat: Deep Emulation for Satellite Data Mining Frontiers in Big Data remote sensing machine learning deep learning atmospheric correction emulator |
author_facet |
Kate Duffy Kate Duffy Thomas Vandal Thomas Vandal Shuang Li Shuang Li Sangram Ganguly Sangram Ganguly Ramakrishna Nemani Auroop R. Ganguly |
author_sort |
Kate Duffy |
title |
DeepEmSat: Deep Emulation for Satellite Data Mining |
title_short |
DeepEmSat: Deep Emulation for Satellite Data Mining |
title_full |
DeepEmSat: Deep Emulation for Satellite Data Mining |
title_fullStr |
DeepEmSat: Deep Emulation for Satellite Data Mining |
title_full_unstemmed |
DeepEmSat: Deep Emulation for Satellite Data Mining |
title_sort |
deepemsat: deep emulation for satellite data mining |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Big Data |
issn |
2624-909X |
publishDate |
2019-12-01 |
description |
The growing volume of Earth science data available from climate simulations and satellite remote sensing offers unprecedented opportunity for scientific insight, while also presenting computational challenges. One potential area of impact is atmospheric correction, where physics-based numerical models retrieve surface reflectance information from top of atmosphere observations, and are too computationally intensive to be run in real time. Machine learning methods have demonstrated potential as fast statistical models for expensive simulations and for extracting credible insights from complex datasets. Here, we develop DeepEmSat: a deep learning emulator approach for atmospheric correction, and offer comparison against physics-based models to support the hypothesis that deep learning can make a contribution to the efficient processing of satellite images. |
topic |
remote sensing machine learning deep learning atmospheric correction emulator |
url |
https://www.frontiersin.org/article/10.3389/fdata.2019.00042/full |
work_keys_str_mv |
AT kateduffy deepemsatdeepemulationforsatellitedatamining AT kateduffy deepemsatdeepemulationforsatellitedatamining AT thomasvandal deepemsatdeepemulationforsatellitedatamining AT thomasvandal deepemsatdeepemulationforsatellitedatamining AT shuangli deepemsatdeepemulationforsatellitedatamining AT shuangli deepemsatdeepemulationforsatellitedatamining AT sangramganguly deepemsatdeepemulationforsatellitedatamining AT sangramganguly deepemsatdeepemulationforsatellitedatamining AT ramakrishnanemani deepemsatdeepemulationforsatellitedatamining AT aurooprganguly deepemsatdeepemulationforsatellitedatamining |
_version_ |
1724838215515373568 |