Application of a Convolutional Neural Network for the Detection of Sea Ice Leads

Despite accounting for a small fraction of the surface area in the Arctic, long and narrow sea ice fractures, known as “leads”, play a critical role in the energy flux between the ocean and atmosphere. As the volume of sea ice in the Arctic has declined over the past few decades, it is increasingly...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jay P. Hoffman, Steven A. Ackerman, Yinghui Liu, Jeffrey R. Key, Iain L. McConnell
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	leads sea ice MODIS VIIRS convolutional neural network U-Net Science Q
Acceso en línea:	https://doaj.org/article/274f2cebf4bd43319714c2b2b21f6e83
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:274f2cebf4bd43319714c2b2b21f6e83
record_format	dspace
spelling	oai:doaj.org-article:274f2cebf4bd43319714c2b2b21f6e832021-11-25T18:54:27ZApplication of a Convolutional Neural Network for the Detection of Sea Ice Leads10.3390/rs132245712072-4292https://doaj.org/article/274f2cebf4bd43319714c2b2b21f6e832021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4571https://doaj.org/toc/2072-4292Despite accounting for a small fraction of the surface area in the Arctic, long and narrow sea ice fractures, known as “leads”, play a critical role in the energy flux between the ocean and atmosphere. As the volume of sea ice in the Arctic has declined over the past few decades, it is increasingly important to monitor the corresponding changes in sea ice leads. A novel approach has been developed using artificial intelligence (AI) to detect sea ice leads using satellite thermal infrared window data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS). In this new approach, a particular type of convolutional neural network, a U-Net, replaces a series of conventional image processing tests from our legacy algorithm. Results show the new approach has a high detection accuracy with F1 Scores on the order of 0.7. Compared to the legacy algorithm, the new algorithm shows improvement, with more true positives, fewer false positives, fewer false negatives, and better agreement between satellite instruments.Jay P. HoffmanSteven A. AckermanYinghui LiuJeffrey R. KeyIain L. McConnellMDPI AGarticleleadssea iceMODISVIIRSconvolutional neural networkU-NetScienceQENRemote Sensing, Vol 13, Iss 4571, p 4571 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	leads sea ice MODIS VIIRS convolutional neural network U-Net Science Q
spellingShingle	leads sea ice MODIS VIIRS convolutional neural network U-Net Science Q Jay P. Hoffman Steven A. Ackerman Yinghui Liu Jeffrey R. Key Iain L. McConnell Application of a Convolutional Neural Network for the Detection of Sea Ice Leads
description	Despite accounting for a small fraction of the surface area in the Arctic, long and narrow sea ice fractures, known as “leads”, play a critical role in the energy flux between the ocean and atmosphere. As the volume of sea ice in the Arctic has declined over the past few decades, it is increasingly important to monitor the corresponding changes in sea ice leads. A novel approach has been developed using artificial intelligence (AI) to detect sea ice leads using satellite thermal infrared window data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS). In this new approach, a particular type of convolutional neural network, a U-Net, replaces a series of conventional image processing tests from our legacy algorithm. Results show the new approach has a high detection accuracy with F1 Scores on the order of 0.7. Compared to the legacy algorithm, the new algorithm shows improvement, with more true positives, fewer false positives, fewer false negatives, and better agreement between satellite instruments.
format	article
author	Jay P. Hoffman Steven A. Ackerman Yinghui Liu Jeffrey R. Key Iain L. McConnell
author_facet	Jay P. Hoffman Steven A. Ackerman Yinghui Liu Jeffrey R. Key Iain L. McConnell
author_sort	Jay P. Hoffman
title	Application of a Convolutional Neural Network for the Detection of Sea Ice Leads
title_short	Application of a Convolutional Neural Network for the Detection of Sea Ice Leads
title_full	Application of a Convolutional Neural Network for the Detection of Sea Ice Leads
title_fullStr	Application of a Convolutional Neural Network for the Detection of Sea Ice Leads
title_full_unstemmed	Application of a Convolutional Neural Network for the Detection of Sea Ice Leads
title_sort	application of a convolutional neural network for the detection of sea ice leads
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/274f2cebf4bd43319714c2b2b21f6e83
work_keys_str_mv	AT jayphoffman applicationofaconvolutionalneuralnetworkforthedetectionofseaiceleads AT stevenaackerman applicationofaconvolutionalneuralnetworkforthedetectionofseaiceleads AT yinghuiliu applicationofaconvolutionalneuralnetworkforthedetectionofseaiceleads AT jeffreyrkey applicationofaconvolutionalneuralnetworkforthedetectionofseaiceleads AT iainlmcconnell applicationofaconvolutionalneuralnetworkforthedetectionofseaiceleads
_version_	1718410575477735424

Application of a Convolutional Neural Network for the Detection of Sea Ice Leads

Ejemplares similares