The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method
When navigating ships in cold regions, sea ice concentration plays a crucial role in determining a ship’s navigability. However, automatically extracting the sea ice concentration and floe size distribution remains challenging, due to the difficulty in detecting all the ice floes from the images cap...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2023
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs15102663 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/15/10/2663/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/15/10/2663/ 2023-08-20T04:09:41+02:00 The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method Li Zhou Jinyan Cai Shifeng Ding agris 2023-05-19 application/pdf https://doi.org/10.3390/rs15102663 EN eng Multidisciplinary Digital Publishing Institute Ocean Remote Sensing https://dx.doi.org/10.3390/rs15102663 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 15; Issue 10; Pages: 2663 deep learning ice floe identification image processing floe size distribution ice concentration YOLACT Text 2023 ftmdpi https://doi.org/10.3390/rs15102663 2023-08-01T10:08:43Z When navigating ships in cold regions, sea ice concentration plays a crucial role in determining a ship’s navigability. However, automatically extracting the sea ice concentration and floe size distribution remains challenging, due to the difficulty in detecting all the ice floes from the images captured in complex polar environments, particularly those that include both ships and sea ice. In this paper, we propose using the YOLACT network to address this issue. Cameras installed on the ship collect images during transit and an image dataset is constructed to train a model that can intelligently identify all the targets in the image and remove any noisy targets. To overcome the challenge of identifying seemingly connected ice floes, the non-maximum suppression (NMS) in YOLACT is improved. Binarization is then applied to process the detection results, with the aim of obtaining an accurate sea ice concentration. We present a color map and histogram of the associated floe size distribution based on the ice size. The speed of calculating the sea ice density of each image reaches 21 FPS and the results show that sea ice concentration and floe size distribution can be accurately measured. We provide a case study to demonstrate the effectiveness of the proposed approach. Text Sea ice MDPI Open Access Publishing Remote Sensing 15 10 2663 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
deep learning ice floe identification image processing floe size distribution ice concentration YOLACT |
spellingShingle |
deep learning ice floe identification image processing floe size distribution ice concentration YOLACT Li Zhou Jinyan Cai Shifeng Ding The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method |
topic_facet |
deep learning ice floe identification image processing floe size distribution ice concentration YOLACT |
description |
When navigating ships in cold regions, sea ice concentration plays a crucial role in determining a ship’s navigability. However, automatically extracting the sea ice concentration and floe size distribution remains challenging, due to the difficulty in detecting all the ice floes from the images captured in complex polar environments, particularly those that include both ships and sea ice. In this paper, we propose using the YOLACT network to address this issue. Cameras installed on the ship collect images during transit and an image dataset is constructed to train a model that can intelligently identify all the targets in the image and remove any noisy targets. To overcome the challenge of identifying seemingly connected ice floes, the non-maximum suppression (NMS) in YOLACT is improved. Binarization is then applied to process the detection results, with the aim of obtaining an accurate sea ice concentration. We present a color map and histogram of the associated floe size distribution based on the ice size. The speed of calculating the sea ice density of each image reaches 21 FPS and the results show that sea ice concentration and floe size distribution can be accurately measured. We provide a case study to demonstrate the effectiveness of the proposed approach. |
format |
Text |
author |
Li Zhou Jinyan Cai Shifeng Ding |
author_facet |
Li Zhou Jinyan Cai Shifeng Ding |
author_sort |
Li Zhou |
title |
The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method |
title_short |
The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method |
title_full |
The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method |
title_fullStr |
The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method |
title_full_unstemmed |
The Identification of Ice Floes and Calculation of Sea Ice Concentration Based on a Deep Learning Method |
title_sort |
identification of ice floes and calculation of sea ice concentration based on a deep learning method |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2023 |
url |
https://doi.org/10.3390/rs15102663 |
op_coverage |
agris |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
Remote Sensing; Volume 15; Issue 10; Pages: 2663 |
op_relation |
Ocean Remote Sensing https://dx.doi.org/10.3390/rs15102663 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs15102663 |
container_title |
Remote Sensing |
container_volume |
15 |
container_issue |
10 |
container_start_page |
2663 |
_version_ |
1774723297584349184 |