Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks
An accurate algorithm for sea ice segmentation is critical for monitoring sea ice parameters of ship navigation in ice-covered seas, as it can automatically extract ice objects and corresponding information to compute essential parameters such as surface ice concentration and ice floe size. In this...
Published in: | International Journal of Applied Earth Observation and Geoinformation |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Elsevier
2022
|
Subjects: | |
Online Access: | https://doi.org/10.1016/j.jag.2022.102885 https://doaj.org/article/9dfd37300945490da5b7360d027c07bb |
id |
ftdoajarticles:oai:doaj.org/article:9dfd37300945490da5b7360d027c07bb |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:9dfd37300945490da5b7360d027c07bb 2023-05-15T18:16:45+02:00 Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks Chengqian Zhang Xiaodong Chen Shunying Ji 2022-08-01T00:00:00Z https://doi.org/10.1016/j.jag.2022.102885 https://doaj.org/article/9dfd37300945490da5b7360d027c07bb EN eng Elsevier http://www.sciencedirect.com/science/article/pii/S1569843222000875 https://doaj.org/toc/1569-8432 1569-8432 doi:10.1016/j.jag.2022.102885 https://doaj.org/article/9dfd37300945490da5b7360d027c07bb International Journal of Applied Earth Observations and Geoinformation, Vol 112, Iss , Pp 102885- (2022) Semantic sea-ice image segmentation Deep convolutional neural networks Multi-scale features Attention module Physical geography GB3-5030 Environmental sciences GE1-350 article 2022 ftdoajarticles https://doi.org/10.1016/j.jag.2022.102885 2022-12-30T20:38:44Z An accurate algorithm for sea ice segmentation is critical for monitoring sea ice parameters of ship navigation in ice-covered seas, as it can automatically extract ice objects and corresponding information to compute essential parameters such as surface ice concentration and ice floe size. In this paper, based on digital images captured by onboard cameras, a novel network called Ice-Deeplab for pixel-wise ice image segmentation is proposed. The Ice-Deeplab network is constructed using the deep convolutional neural network Deeplab and is modified with an attention module and an improved decoding structure. To investigate its reliability, the Ice-Deeplab network is applied to a 320-image dataset, with 80% for training and 20% for validation. The experiments demonstrated that the proposed Ice-Deeplab yields better segmentation results than the original Deeplab model under different validation scenarios, achieving an overall accuracy of 90.5% among the classes sea-ice, ocean, and sky. Moreover, the proposed model was applied to un-labelled test data to demonstrate its generalisation ability for real-time ice segmentation. Article in Journal/Newspaper Sea ice Directory of Open Access Journals: DOAJ Articles International Journal of Applied Earth Observation and Geoinformation 112 102885 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Semantic sea-ice image segmentation Deep convolutional neural networks Multi-scale features Attention module Physical geography GB3-5030 Environmental sciences GE1-350 |
spellingShingle |
Semantic sea-ice image segmentation Deep convolutional neural networks Multi-scale features Attention module Physical geography GB3-5030 Environmental sciences GE1-350 Chengqian Zhang Xiaodong Chen Shunying Ji Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
topic_facet |
Semantic sea-ice image segmentation Deep convolutional neural networks Multi-scale features Attention module Physical geography GB3-5030 Environmental sciences GE1-350 |
description |
An accurate algorithm for sea ice segmentation is critical for monitoring sea ice parameters of ship navigation in ice-covered seas, as it can automatically extract ice objects and corresponding information to compute essential parameters such as surface ice concentration and ice floe size. In this paper, based on digital images captured by onboard cameras, a novel network called Ice-Deeplab for pixel-wise ice image segmentation is proposed. The Ice-Deeplab network is constructed using the deep convolutional neural network Deeplab and is modified with an attention module and an improved decoding structure. To investigate its reliability, the Ice-Deeplab network is applied to a 320-image dataset, with 80% for training and 20% for validation. The experiments demonstrated that the proposed Ice-Deeplab yields better segmentation results than the original Deeplab model under different validation scenarios, achieving an overall accuracy of 90.5% among the classes sea-ice, ocean, and sky. Moreover, the proposed model was applied to un-labelled test data to demonstrate its generalisation ability for real-time ice segmentation. |
format |
Article in Journal/Newspaper |
author |
Chengqian Zhang Xiaodong Chen Shunying Ji |
author_facet |
Chengqian Zhang Xiaodong Chen Shunying Ji |
author_sort |
Chengqian Zhang |
title |
Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
title_short |
Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
title_full |
Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
title_fullStr |
Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
title_full_unstemmed |
Semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
title_sort |
semantic image segmentation for sea ice parameters recognition using deep convolutional neural networks |
publisher |
Elsevier |
publishDate |
2022 |
url |
https://doi.org/10.1016/j.jag.2022.102885 https://doaj.org/article/9dfd37300945490da5b7360d027c07bb |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
International Journal of Applied Earth Observations and Geoinformation, Vol 112, Iss , Pp 102885- (2022) |
op_relation |
http://www.sciencedirect.com/science/article/pii/S1569843222000875 https://doaj.org/toc/1569-8432 1569-8432 doi:10.1016/j.jag.2022.102885 https://doaj.org/article/9dfd37300945490da5b7360d027c07bb |
op_doi |
https://doi.org/10.1016/j.jag.2022.102885 |
container_title |
International Journal of Applied Earth Observation and Geoinformation |
container_volume |
112 |
container_start_page |
102885 |
_version_ |
1766190609376215040 |