Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches

Precise coastal shoreline mapping is essential for monitoring changes in erosion rates, surface hydrology, and ecosystem structure and function. Monitoring water bodies in the Arctic National Wildlife Refuge (ANWR) is of high importance, especially considering the potential for oil and natural gas e...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Bibek Aryal, Stephen M. Escarzaga, Sergio A. Vargas Zesati, Miguel Velez-Reyes, Olac Fuentes, Craig Tweedie
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	land water segmentation remote sensing deep learning sparse labels Science Q
Acceso en línea:	https://doaj.org/article/72db7bdf46df409681f6ce7218ef9a4b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:72db7bdf46df409681f6ce7218ef9a4b
record_format	dspace
spelling	oai:doaj.org-article:72db7bdf46df409681f6ce7218ef9a4b2021-11-25T18:54:28ZSemi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches10.3390/rs132245722072-4292https://doaj.org/article/72db7bdf46df409681f6ce7218ef9a4b2021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4572https://doaj.org/toc/2072-4292Precise coastal shoreline mapping is essential for monitoring changes in erosion rates, surface hydrology, and ecosystem structure and function. Monitoring water bodies in the Arctic National Wildlife Refuge (ANWR) is of high importance, especially considering the potential for oil and natural gas exploration in the region. In this work, we propose a modified variant of the Deep Neural Network based U-Net Architecture for the automated mapping of 4 Band Orthorectified NOAA Airborne Imagery using sparsely labeled training data and compare it to the performance of traditional Machine Learning (ML) based approaches—namely, random forest, xgboost—and spectral water indices—Normalized Difference Water Index (NDWI), and Normalized Difference Surface Water Index (NDSWI)—to support shoreline mapping of Arctic coastlines. We conclude that it is possible to modify the U-Net model to accept sparse labels as input and the results are comparable to other ML methods (an Intersection-over-Union (IoU) of 94.86% using U-Net vs. an IoU of 95.05% using the best performing method).Bibek AryalStephen M. EscarzagaSergio A. Vargas ZesatiMiguel Velez-ReyesOlac FuentesCraig TweedieMDPI AGarticleland water segmentationremote sensingdeep learningsparse labelsScienceQENRemote Sensing, Vol 13, Iss 4572, p 4572 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	land water segmentation remote sensing deep learning sparse labels Science Q
spellingShingle	land water segmentation remote sensing deep learning sparse labels Science Q Bibek Aryal Stephen M. Escarzaga Sergio A. Vargas Zesati Miguel Velez-Reyes Olac Fuentes Craig Tweedie Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches
description	Precise coastal shoreline mapping is essential for monitoring changes in erosion rates, surface hydrology, and ecosystem structure and function. Monitoring water bodies in the Arctic National Wildlife Refuge (ANWR) is of high importance, especially considering the potential for oil and natural gas exploration in the region. In this work, we propose a modified variant of the Deep Neural Network based U-Net Architecture for the automated mapping of 4 Band Orthorectified NOAA Airborne Imagery using sparsely labeled training data and compare it to the performance of traditional Machine Learning (ML) based approaches—namely, random forest, xgboost—and spectral water indices—Normalized Difference Water Index (NDWI), and Normalized Difference Surface Water Index (NDSWI)—to support shoreline mapping of Arctic coastlines. We conclude that it is possible to modify the U-Net model to accept sparse labels as input and the results are comparable to other ML methods (an Intersection-over-Union (IoU) of 94.86% using U-Net vs. an IoU of 95.05% using the best performing method).
format	article
author	Bibek Aryal Stephen M. Escarzaga Sergio A. Vargas Zesati Miguel Velez-Reyes Olac Fuentes Craig Tweedie
author_facet	Bibek Aryal Stephen M. Escarzaga Sergio A. Vargas Zesati Miguel Velez-Reyes Olac Fuentes Craig Tweedie
author_sort	Bibek Aryal
title	Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches
title_short	Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches
title_full	Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches
title_fullStr	Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches
title_full_unstemmed	Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches
title_sort	semi-automated semantic segmentation of arctic shorelines using very high-resolution airborne imagery, spectral indices and weakly supervised machine learning approaches
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/72db7bdf46df409681f6ce7218ef9a4b
work_keys_str_mv	AT bibekaryal semiautomatedsemanticsegmentationofarcticshorelinesusingveryhighresolutionairborneimageryspectralindicesandweaklysupervisedmachinelearningapproaches AT stephenmescarzaga semiautomatedsemanticsegmentationofarcticshorelinesusingveryhighresolutionairborneimageryspectralindicesandweaklysupervisedmachinelearningapproaches AT sergioavargaszesati semiautomatedsemanticsegmentationofarcticshorelinesusingveryhighresolutionairborneimageryspectralindicesandweaklysupervisedmachinelearningapproaches AT miguelvelezreyes semiautomatedsemanticsegmentationofarcticshorelinesusingveryhighresolutionairborneimageryspectralindicesandweaklysupervisedmachinelearningapproaches AT olacfuentes semiautomatedsemanticsegmentationofarcticshorelinesusingveryhighresolutionairborneimageryspectralindicesandweaklysupervisedmachinelearningapproaches AT craigtweedie semiautomatedsemanticsegmentationofarcticshorelinesusingveryhighresolutionairborneimageryspectralindicesandweaklysupervisedmachinelearningapproaches
_version_	1718410595833741312

Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches

Ejemplares similares