Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data

Maria Luize Pinheiro; Luiz Cortinhas; Cesar Diniz; Raian V. Maretto; Mateus Grellert

doi:10.1109/IGARSS52108.2023.10282540

Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data

Maria Luize Pinheiro^*, Luiz Cortinhas, Cesar Diniz, Raian V. Maretto, Mateus Grellert

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

67 Downloads (Pure)

Abstract

Hypersaline tidal flats are plane areas usually related to mangrove forests, acting as guard and buffer against rising sea levels, and as maintainer of regional biodiversity. Such areas are primarily impacted by anthropogenic and natural activities, such as sea-salt extraction and pollution, so identifying and monitoring them is an important and challenging task. The present work uses a U-shaped Convolutional Neural Network architecture to systematically classify such formations over Landsat imagery. A large dataset containing data from 1985 to 2021 of the Brazilian Coastal Zone is used to train and evaluate our model. Experimental results show that the total area increased by 58.6 km² from 1985 to 2001, and decreased by approximately 92 km² from 2001 to 2021, representing a total reduction of ≈ 33.34 km² for the entire period. We also show that our model outperforms a related solution trained with the same dataset, achieving 70% and 86% for 1985 and 2020 respectively, against 69% and 82%.

Original language	English
Title of host publication	IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings
Publisher	IEEE
Pages	3337-3340
Number of pages	4
ISBN (Electronic)	9798350320107
DOIs	https://doi.org/10.1109/IGARSS52108.2023.10282540
Publication status	Published - 2023
Event	43rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023 - Pasadena Convention Center, Pasadena, United States Duration: 16 Jul 2023 → 21 Jul 2023 Conference number: 43 https://2023.ieeeigarss.org/index.php

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Volume	2023-July

Conference

Conference	43rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023
Abbreviated title	IGARSS 2023
Country/Territory	United States
City	Pasadena
Period	16/07/23 → 21/07/23
Internet address	https://2023.ieeeigarss.org/index.php

Keywords

Deep Learning
Hypersaline tidal flats
Image Segmentation
Neural Network
2024 OA procedure

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/IGARSS52108.2023.10282540

Hypersaline_Tidal_Flats_Detection_Using_Deep_Learning_Over_37_Years_of_Landsat_DataFinal published version, 12.4 MBLicence: Taverne

Cite this

Pinheiro, M. L., Cortinhas, L., Diniz, C., Maretto, R. V., & Grellert, M. (2023). Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data. In IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings (pp. 3337-3340). (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2023-July). IEEE. https://doi.org/10.1109/IGARSS52108.2023.10282540

@inproceedings{7c0112cc5b2348f5afe7c97b9768b08c,

title = "Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data",

abstract = "Hypersaline tidal flats are plane areas usually related to mangrove forests, acting as guard and buffer against rising sea levels, and as maintainer of regional biodiversity. Such areas are primarily impacted by anthropogenic and natural activities, such as sea-salt extraction and pollution, so identifying and monitoring them is an important and challenging task. The present work uses a U-shaped Convolutional Neural Network architecture to systematically classify such formations over Landsat imagery. A large dataset containing data from 1985 to 2021 of the Brazilian Coastal Zone is used to train and evaluate our model. Experimental results show that the total area increased by 58.6 km2 from 1985 to 2001, and decreased by approximately 92 km2 from 2001 to 2021, representing a total reduction of ≈ 33.34 km2 for the entire period. We also show that our model outperforms a related solution trained with the same dataset, achieving 70% and 86% for 1985 and 2020 respectively, against 69% and 82%.",

keywords = "Deep Learning, Hypersaline tidal flats, Image Segmentation, Neural Network, 2024 OA procedure",

author = "Pinheiro, {Maria Luize} and Luiz Cortinhas and Cesar Diniz and Maretto, {Raian V.} and Mateus Grellert",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 43rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023, IGARSS 2023 ; Conference date: 16-07-2023 Through 21-07-2023",

year = "2023",

doi = "10.1109/IGARSS52108.2023.10282540",

language = "English",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

publisher = "IEEE",

pages = "3337--3340",

booktitle = "IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings",

address = "United States",

url = "https://2023.ieeeigarss.org/index.php",

}

Pinheiro, ML, Cortinhas, L, Diniz, C, Maretto, RV & Grellert, M 2023, Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data. in IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. International Geoscience and Remote Sensing Symposium (IGARSS), vol. 2023-July, IEEE, pp. 3337-3340, 43rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023, Pasadena, California, United States, 16/07/23. https://doi.org/10.1109/IGARSS52108.2023.10282540

Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data. / Pinheiro, Maria Luize; Cortinhas, Luiz; Diniz, Cesar et al.
IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. IEEE, 2023. p. 3337-3340 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2023-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data

AU - Pinheiro, Maria Luize

AU - Cortinhas, Luiz

AU - Diniz, Cesar

AU - Maretto, Raian V.

AU - Grellert, Mateus

N1 - Conference code: 43

PY - 2023

Y1 - 2023

N2 - Hypersaline tidal flats are plane areas usually related to mangrove forests, acting as guard and buffer against rising sea levels, and as maintainer of regional biodiversity. Such areas are primarily impacted by anthropogenic and natural activities, such as sea-salt extraction and pollution, so identifying and monitoring them is an important and challenging task. The present work uses a U-shaped Convolutional Neural Network architecture to systematically classify such formations over Landsat imagery. A large dataset containing data from 1985 to 2021 of the Brazilian Coastal Zone is used to train and evaluate our model. Experimental results show that the total area increased by 58.6 km2 from 1985 to 2001, and decreased by approximately 92 km2 from 2001 to 2021, representing a total reduction of ≈ 33.34 km2 for the entire period. We also show that our model outperforms a related solution trained with the same dataset, achieving 70% and 86% for 1985 and 2020 respectively, against 69% and 82%.

AB - Hypersaline tidal flats are plane areas usually related to mangrove forests, acting as guard and buffer against rising sea levels, and as maintainer of regional biodiversity. Such areas are primarily impacted by anthropogenic and natural activities, such as sea-salt extraction and pollution, so identifying and monitoring them is an important and challenging task. The present work uses a U-shaped Convolutional Neural Network architecture to systematically classify such formations over Landsat imagery. A large dataset containing data from 1985 to 2021 of the Brazilian Coastal Zone is used to train and evaluate our model. Experimental results show that the total area increased by 58.6 km2 from 1985 to 2001, and decreased by approximately 92 km2 from 2001 to 2021, representing a total reduction of ≈ 33.34 km2 for the entire period. We also show that our model outperforms a related solution trained with the same dataset, achieving 70% and 86% for 1985 and 2020 respectively, against 69% and 82%.

KW - Deep Learning

KW - Hypersaline tidal flats

KW - Image Segmentation

KW - Neural Network

KW - 2024 OA procedure

U2 - 10.1109/IGARSS52108.2023.10282540

DO - 10.1109/IGARSS52108.2023.10282540

M3 - Conference contribution

AN - SCOPUS:85178383334

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 3337

EP - 3340

BT - IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings

PB - IEEE

T2 - 43rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023

Y2 - 16 July 2023 through 21 July 2023

ER -

Hypersaline Tidal Flats Detection Using Deep Learning Over 37 Years of Landsat Data

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this