Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks
Sea ice concentration is of great interest to ship navigators and scientists who require regional ice cover understanding. Passive microwave data and image analysis charts are typically used to estimate ice concentration, but these have limitations. Estimates obtained from passive microwave data hav...
Main Author: | |
---|---|
Format: | Master Thesis |
Language: | English |
Published: |
University of Waterloo
2020
|
Subjects: | |
Online Access: | http://hdl.handle.net/10012/16213 |
id |
ftunivwaterloo:oai:uwspace.uwaterloo.ca:10012/16213 |
---|---|
record_format |
openpolar |
spelling |
ftunivwaterloo:oai:uwspace.uwaterloo.ca:10012/16213 2023-05-15T18:17:31+02:00 Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks Radhakrishnan, Keerthijan 2020-08-20 http://hdl.handle.net/10012/16213 en eng University of Waterloo http://hdl.handle.net/10012/16213 sea ice concentration synthetic aperture radar convolutional neural networks fully convolutional networks Master Thesis 2020 ftunivwaterloo 2022-06-18T23:03:00Z Sea ice concentration is of great interest to ship navigators and scientists who require regional ice cover understanding. Passive microwave data and image analysis charts are typically used to estimate ice concentration, but these have limitations. Estimates obtained from passive microwave data have coarse spatial resolution, may be biased due to weather filters that reduce atmospheric contamination, and often perform poorly in marginal ice zones. Image analysis charts are not as precise and subjective to analyst interpretation. Synthetic aperture radar (SAR) images are finer resolution satellite images that can be used to observe oceans. However, the complex interactions between the SAR signal and water and ice make it a difficult process to estimate sea ice concentration. Previous studies have found that deep learning is a viable avenue to estimate ice concentration from SAR images. In these studies, convolutional neural networks (CNNs) have been successful due to their ability to learn spatial features in their convolutional layers. To overcome the shortcomings of ice concentration estimation, we have uniquely implemented a U-net with SAR images as inputs and use estimates obtained from passive microwave data as training labels. The U-net, due to not being sensitive to patch size, is shown to be an improvement over the CNN models used in previous studies. Data augmentation and a mean absolute error (L1) loss function were applied as well as a curriculum learning method that introduces more open water and consolidated ice regions before incorporating marginal ice regions. The key objectives of this study are (a) to overcome shortcomings of using passive microwave data for model training and (b) to improve ice concentration predictions in marginal ice zones. Evaluating with image analysis charts, a mean absolute error of 7.18\% is achieved, which is lower than errors associated with estimation algorithms using passive microwave data alone. Through qualitative analysis, we also show instances where our proposed ... Master Thesis Sea ice University of Waterloo, Canada: Institutional Repository |
institution |
Open Polar |
collection |
University of Waterloo, Canada: Institutional Repository |
op_collection_id |
ftunivwaterloo |
language |
English |
topic |
sea ice concentration synthetic aperture radar convolutional neural networks fully convolutional networks |
spellingShingle |
sea ice concentration synthetic aperture radar convolutional neural networks fully convolutional networks Radhakrishnan, Keerthijan Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks |
topic_facet |
sea ice concentration synthetic aperture radar convolutional neural networks fully convolutional networks |
description |
Sea ice concentration is of great interest to ship navigators and scientists who require regional ice cover understanding. Passive microwave data and image analysis charts are typically used to estimate ice concentration, but these have limitations. Estimates obtained from passive microwave data have coarse spatial resolution, may be biased due to weather filters that reduce atmospheric contamination, and often perform poorly in marginal ice zones. Image analysis charts are not as precise and subjective to analyst interpretation. Synthetic aperture radar (SAR) images are finer resolution satellite images that can be used to observe oceans. However, the complex interactions between the SAR signal and water and ice make it a difficult process to estimate sea ice concentration. Previous studies have found that deep learning is a viable avenue to estimate ice concentration from SAR images. In these studies, convolutional neural networks (CNNs) have been successful due to their ability to learn spatial features in their convolutional layers. To overcome the shortcomings of ice concentration estimation, we have uniquely implemented a U-net with SAR images as inputs and use estimates obtained from passive microwave data as training labels. The U-net, due to not being sensitive to patch size, is shown to be an improvement over the CNN models used in previous studies. Data augmentation and a mean absolute error (L1) loss function were applied as well as a curriculum learning method that introduces more open water and consolidated ice regions before incorporating marginal ice regions. The key objectives of this study are (a) to overcome shortcomings of using passive microwave data for model training and (b) to improve ice concentration predictions in marginal ice zones. Evaluating with image analysis charts, a mean absolute error of 7.18\% is achieved, which is lower than errors associated with estimation algorithms using passive microwave data alone. Through qualitative analysis, we also show instances where our proposed ... |
format |
Master Thesis |
author |
Radhakrishnan, Keerthijan |
author_facet |
Radhakrishnan, Keerthijan |
author_sort |
Radhakrishnan, Keerthijan |
title |
Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks |
title_short |
Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks |
title_full |
Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks |
title_fullStr |
Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks |
title_full_unstemmed |
Sea Ice Concentration Estimation: Using Passive Microwave and SAR Data with Fully Convolutional Networks |
title_sort |
sea ice concentration estimation: using passive microwave and sar data with fully convolutional networks |
publisher |
University of Waterloo |
publishDate |
2020 |
url |
http://hdl.handle.net/10012/16213 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_relation |
http://hdl.handle.net/10012/16213 |
_version_ |
1766191775900237824 |