Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal
Sea-ice identification is an essential process for safety critical navigation support of surface vessels in polar waters. Semantic segmentation has drawn much attention as an enabling technique for fast detection of objects in a scene including sea-ice conditions. Identifying sea-ice is a challengin...
Published in: | IEEE Access |
---|---|
Main Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
IEEE
2022
|
Subjects: | |
Online Access: | https://doi.org/10.1109/ACCESS.2022.3150969 https://doaj.org/article/91f019510305425587be037987d704d8 |
id |
ftdoajarticles:oai:doaj.org/article:91f019510305425587be037987d704d8 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:91f019510305425587be037987d704d8 2023-05-15T18:16:25+02:00 Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal Nahed M. Alsharay Yuanzhu Chen Octavia A. Dobre Oscar De Silva 2022-01-01T00:00:00Z https://doi.org/10.1109/ACCESS.2022.3150969 https://doaj.org/article/91f019510305425587be037987d704d8 EN eng IEEE https://ieeexplore.ieee.org/document/9709781/ https://doaj.org/toc/2169-3536 2169-3536 doi:10.1109/ACCESS.2022.3150969 https://doaj.org/article/91f019510305425587be037987d704d8 IEEE Access, Vol 10, Pp 21599-21607 (2022) Convolutional neural networks raindrop removing sea-ice semantic segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971 article 2022 ftdoajarticles https://doi.org/10.1109/ACCESS.2022.3150969 2022-12-31T16:15:04Z Sea-ice identification is an essential process for safety critical navigation support of surface vessels in polar waters. Semantic segmentation has drawn much attention as an enabling technique for fast detection of objects in a scene including sea-ice conditions. Identifying sea-ice is a challenging problem, especially in the presence of raindrops. The raindrop alters the boundaries of the objects in the scene, and thus, degrades the identification performance. In this work, a raindrop removing framework is developed to enhance the classification performance. Three deep-learning semantic segmentation networks are trained to classify the scene of sea-ice images into ice, water, ship, and sky. The deep-learning networks are VGG-16, fully convolutional network, and pyramid scene parsing network. Transfer learning along with data augmentation operations have been implemented to improve the training process. Results illustrate that data augmentation operations enhance the performance of the three models. Moreover, the raindrop removing framework improves the models’ performance, e.g. the average intersection over union of the VGG-16 model is improved from 85.91% to 91.70%. Article in Journal/Newspaper Sea ice Directory of Open Access Journals: DOAJ Articles Pyramid ENVELOPE(157.300,157.300,-81.333,-81.333) IEEE Access 10 21599 21607 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Convolutional neural networks raindrop removing sea-ice semantic segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971 |
spellingShingle |
Convolutional neural networks raindrop removing sea-ice semantic segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971 Nahed M. Alsharay Yuanzhu Chen Octavia A. Dobre Oscar De Silva Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal |
topic_facet |
Convolutional neural networks raindrop removing sea-ice semantic segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971 |
description |
Sea-ice identification is an essential process for safety critical navigation support of surface vessels in polar waters. Semantic segmentation has drawn much attention as an enabling technique for fast detection of objects in a scene including sea-ice conditions. Identifying sea-ice is a challenging problem, especially in the presence of raindrops. The raindrop alters the boundaries of the objects in the scene, and thus, degrades the identification performance. In this work, a raindrop removing framework is developed to enhance the classification performance. Three deep-learning semantic segmentation networks are trained to classify the scene of sea-ice images into ice, water, ship, and sky. The deep-learning networks are VGG-16, fully convolutional network, and pyramid scene parsing network. Transfer learning along with data augmentation operations have been implemented to improve the training process. Results illustrate that data augmentation operations enhance the performance of the three models. Moreover, the raindrop removing framework improves the models’ performance, e.g. the average intersection over union of the VGG-16 model is improved from 85.91% to 91.70%. |
format |
Article in Journal/Newspaper |
author |
Nahed M. Alsharay Yuanzhu Chen Octavia A. Dobre Oscar De Silva |
author_facet |
Nahed M. Alsharay Yuanzhu Chen Octavia A. Dobre Oscar De Silva |
author_sort |
Nahed M. Alsharay |
title |
Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal |
title_short |
Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal |
title_full |
Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal |
title_fullStr |
Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal |
title_full_unstemmed |
Improved Sea-Ice Identification Using Semantic Segmentation With Raindrop Removal |
title_sort |
improved sea-ice identification using semantic segmentation with raindrop removal |
publisher |
IEEE |
publishDate |
2022 |
url |
https://doi.org/10.1109/ACCESS.2022.3150969 https://doaj.org/article/91f019510305425587be037987d704d8 |
long_lat |
ENVELOPE(157.300,157.300,-81.333,-81.333) |
geographic |
Pyramid |
geographic_facet |
Pyramid |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
IEEE Access, Vol 10, Pp 21599-21607 (2022) |
op_relation |
https://ieeexplore.ieee.org/document/9709781/ https://doaj.org/toc/2169-3536 2169-3536 doi:10.1109/ACCESS.2022.3150969 https://doaj.org/article/91f019510305425587be037987d704d8 |
op_doi |
https://doi.org/10.1109/ACCESS.2022.3150969 |
container_title |
IEEE Access |
container_volume |
10 |
container_start_page |
21599 |
op_container_end_page |
21607 |
_version_ |
1766190009241567232 |