Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series
The automatic detection of Center Pivot Irrigation Systems (CPIS) is fundamental for establishing public policies, especially in countries with a growth perspective in this technology, like Brazil. Previous studies to detect CPIS using deep learning used single-date optical images, containing limita...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2021-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9513599/ |
id |
doaj-f720753769e14802a5f818c9db6a786f |
---|---|
record_format |
Article |
spelling |
doaj-f720753769e14802a5f818c9db6a786f2021-09-08T23:00:10ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01148447845710.1109/JSTARS.2021.31047269513599Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time SeriesAnesmar Olino de Albuquerque0https://orcid.org/0000-0003-1561-7583Osmar Luiz Ferreira de Carvalho1https://orcid.org/0000-0002-5619-8525Cristiano Rosa e Silva2https://orcid.org/0000-0003-1189-3337Argelica Saiaka Luiz3https://orcid.org/0000-0003-2738-465XPablo P. de Bem4https://orcid.org/0000-0003-3868-8704Roberto Arnaldo Trancoso Gomes5https://orcid.org/0000-0003-4724-4064Renato Fontes Guimaraes6https://orcid.org/0000-0002-9555-043XOsmar Abilio de Carvalho Junior7https://orcid.org/0000-0002-0346-1684Department of Geography, University of Brasilia, Brasilia, BrazilDepartment of Computer Science, University of Brasilia, Brasilia, BrazilDepartment of Geography, University of Brasilia, Brasilia, BrazilDepartment of Geography, University of Brasilia, Brasilia, BrazilDepartment of Geography, University of Brasilia, Brasilia, BrazilDepartment of Geography, University of Brasilia, Brasilia, BrazilDepartment of Geography, University of Brasilia, Brasilia, BrazilDepartment of Geography, University of Brasilia, Brasilia, BrazilThe automatic detection of Center Pivot Irrigation Systems (CPIS) is fundamental for establishing public policies, especially in countries with a growth perspective in this technology, like Brazil. Previous studies to detect CPIS using deep learning used single-date optical images, containing limitations due to seasonal changes and cloud cover. Therefore, this research aimed to detect CPIS using Sentinel-2 multitemporal images (containing six dates) and instance segmentation, considering seasonal variations and different proportions of cloudy images, generalizing the models to detect CPIS in diverse situations. We used a novel augmentation strategy, in which, for each iteration, six images were randomly selected from the time series (from a total of 11 dates) in random order. We evaluated the Mask-RCNN model with the ResNext-101 backbone considering the COCO metrics on six testing sets with different ratios of cloudless (<inline-formula><tex-math notation="LaTeX">$< 20\%$</tex-math></inline-formula>) and cloudy images (<inline-formula><tex-math notation="LaTeX">$>75\%$</tex-math></inline-formula>), from six cloudless images and zero cloudy images (6:0) up to one cloudless image and five cloudy images (1:5). We found that using six cloudless images provided the best metrics [80% average precision (AP), 93% AP with a 0.5 intersection over union threshold (AP50)]. However, results were similar (74% AP, 88% AP50) even in extreme scenarios with abundant cloud presence (1:5 ratio). Our method provides a more adaptive and automatic way to map CPIS from time series, significantly reducing interference such as cloud cover, atmospheric effects, shadow, missing data, and lack of contrast with the surrounding vegetation.https://ieeexplore.ieee.org/document/9513599/Clouddeep learningmask R-CNNtime series |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Anesmar Olino de Albuquerque Osmar Luiz Ferreira de Carvalho Cristiano Rosa e Silva Argelica Saiaka Luiz Pablo P. de Bem Roberto Arnaldo Trancoso Gomes Renato Fontes Guimaraes Osmar Abilio de Carvalho Junior |
spellingShingle |
Anesmar Olino de Albuquerque Osmar Luiz Ferreira de Carvalho Cristiano Rosa e Silva Argelica Saiaka Luiz Pablo P. de Bem Roberto Arnaldo Trancoso Gomes Renato Fontes Guimaraes Osmar Abilio de Carvalho Junior Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Cloud deep learning mask R-CNN time series |
author_facet |
Anesmar Olino de Albuquerque Osmar Luiz Ferreira de Carvalho Cristiano Rosa e Silva Argelica Saiaka Luiz Pablo P. de Bem Roberto Arnaldo Trancoso Gomes Renato Fontes Guimaraes Osmar Abilio de Carvalho Junior |
author_sort |
Anesmar Olino de Albuquerque |
title |
Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series |
title_short |
Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series |
title_full |
Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series |
title_fullStr |
Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series |
title_full_unstemmed |
Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series |
title_sort |
dealing with clouds and seasonal changes for center pivot irrigation systems detection using instance segmentation in sentinel-2 time series |
publisher |
IEEE |
series |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
issn |
2151-1535 |
publishDate |
2021-01-01 |
description |
The automatic detection of Center Pivot Irrigation Systems (CPIS) is fundamental for establishing public policies, especially in countries with a growth perspective in this technology, like Brazil. Previous studies to detect CPIS using deep learning used single-date optical images, containing limitations due to seasonal changes and cloud cover. Therefore, this research aimed to detect CPIS using Sentinel-2 multitemporal images (containing six dates) and instance segmentation, considering seasonal variations and different proportions of cloudy images, generalizing the models to detect CPIS in diverse situations. We used a novel augmentation strategy, in which, for each iteration, six images were randomly selected from the time series (from a total of 11 dates) in random order. We evaluated the Mask-RCNN model with the ResNext-101 backbone considering the COCO metrics on six testing sets with different ratios of cloudless (<inline-formula><tex-math notation="LaTeX">$< 20\%$</tex-math></inline-formula>) and cloudy images (<inline-formula><tex-math notation="LaTeX">$>75\%$</tex-math></inline-formula>), from six cloudless images and zero cloudy images (6:0) up to one cloudless image and five cloudy images (1:5). We found that using six cloudless images provided the best metrics [80% average precision (AP), 93% AP with a 0.5 intersection over union threshold (AP50)]. However, results were similar (74% AP, 88% AP50) even in extreme scenarios with abundant cloud presence (1:5 ratio). Our method provides a more adaptive and automatic way to map CPIS from time series, significantly reducing interference such as cloud cover, atmospheric effects, shadow, missing data, and lack of contrast with the surrounding vegetation. |
topic |
Cloud deep learning mask R-CNN time series |
url |
https://ieeexplore.ieee.org/document/9513599/ |
work_keys_str_mv |
AT anesmarolinodealbuquerque dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT osmarluizferreiradecarvalho dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT cristianorosaesilva dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT argelicasaiakaluiz dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT pablopdebem dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT robertoarnaldotrancosogomes dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT renatofontesguimaraes dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries AT osmarabiliodecarvalhojunior dealingwithcloudsandseasonalchangesforcenterpivotirrigationsystemsdetectionusinginstancesegmentationinsentinel2timeseries |
_version_ |
1717761670685655040 |