Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region
This thesis presents a deep learning tool able to identify ice in radar images fromthe sea-ice environment of the Twhaites glacier outlet. The project is motivatedby the threatening situation of the Thwaites glacier that has been increasingits mass loss rate during the last decade. This is of concer...
Main Author: | |
---|---|
Format: | Bachelor Thesis |
Language: | English |
Published: |
Luleå tekniska universitet, Rymdteknik
2022
|
Subjects: | |
Online Access: | http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-95352 |
id |
ftluleatu:oai:DiVA.org:ltu-95352 |
---|---|
record_format |
openpolar |
spelling |
ftluleatu:oai:DiVA.org:ltu-95352 2023-05-15T14:03:11+02:00 Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region Fuentes Soria, Carmen 2022 application/pdf http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-95352 eng eng Luleå tekniska universitet, Rymdteknik Aalto university http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-95352 info:eu-repo/semantics/openAccess Synthetic Aperture Radar (SAR) Machine Learning (ML) Semantic segmentation Convolutional Neural Networks (CNN) Thwaites Glacier Aerospace Engineering Rymd- och flygteknik Student thesis info:eu-repo/semantics/bachelorThesis text 2022 ftluleatu 2023-02-02T17:10:34Z This thesis presents a deep learning tool able to identify ice in radar images fromthe sea-ice environment of the Twhaites glacier outlet. The project is motivatedby the threatening situation of the Thwaites glacier that has been increasingits mass loss rate during the last decade. This is of concern considering thelarge mass of ice held by the glacier, that in case of melting, could increasethe mean sea level by more than +65 cm [1]. The algorithm generated alongthis work is intended to help in the generation of navigation charts and identificationof icebergs in future stages of the project, outside of the scope of this thesis.The data used for this task are ICEYE’s X-band radar images from the Thwaitessea-ice environment, the target area to be studied. The corresponding groundtruth for each of the samples has been manually generated identifying the iceand icebergs present in each image. Additional data processing includes tiling,to increment the number of samples, and augmentation, done by horizontal andvertical flips of a random number of tiles.The proposed tool performs semantic segmentation on radar images classifyingthe class "Ice". It is developed by a deep learning Convolutional Neural Network(CNN) model, trained with prepared ICEYE’s radar images. The model reachesvalues of F1 metric higher than 89% in the images of the target area (Thwaitessea-ice environment) and is able to generalize to different regions of Antarctica,reaching values of F1 = 80 %. A potential alternative version of the algorithm isproposed and discussed. This alternative score F1 values higher than F1 > 95 %for images of the target environment and F1 = 87 % for the image of the differentregion. However, it must not be confirmed as the final algorithm due to the needfor further verification. Bachelor Thesis Antarc* Antarctica Sea ice Thwaites Glacier Luleå University of Technology Publications (DiVA) Thwaites Glacier ENVELOPE(-106.750,-106.750,-75.500,-75.500) |
institution |
Open Polar |
collection |
Luleå University of Technology Publications (DiVA) |
op_collection_id |
ftluleatu |
language |
English |
topic |
Synthetic Aperture Radar (SAR) Machine Learning (ML) Semantic segmentation Convolutional Neural Networks (CNN) Thwaites Glacier Aerospace Engineering Rymd- och flygteknik |
spellingShingle |
Synthetic Aperture Radar (SAR) Machine Learning (ML) Semantic segmentation Convolutional Neural Networks (CNN) Thwaites Glacier Aerospace Engineering Rymd- och flygteknik Fuentes Soria, Carmen Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region |
topic_facet |
Synthetic Aperture Radar (SAR) Machine Learning (ML) Semantic segmentation Convolutional Neural Networks (CNN) Thwaites Glacier Aerospace Engineering Rymd- och flygteknik |
description |
This thesis presents a deep learning tool able to identify ice in radar images fromthe sea-ice environment of the Twhaites glacier outlet. The project is motivatedby the threatening situation of the Thwaites glacier that has been increasingits mass loss rate during the last decade. This is of concern considering thelarge mass of ice held by the glacier, that in case of melting, could increasethe mean sea level by more than +65 cm [1]. The algorithm generated alongthis work is intended to help in the generation of navigation charts and identificationof icebergs in future stages of the project, outside of the scope of this thesis.The data used for this task are ICEYE’s X-band radar images from the Thwaitessea-ice environment, the target area to be studied. The corresponding groundtruth for each of the samples has been manually generated identifying the iceand icebergs present in each image. Additional data processing includes tiling,to increment the number of samples, and augmentation, done by horizontal andvertical flips of a random number of tiles.The proposed tool performs semantic segmentation on radar images classifyingthe class "Ice". It is developed by a deep learning Convolutional Neural Network(CNN) model, trained with prepared ICEYE’s radar images. The model reachesvalues of F1 metric higher than 89% in the images of the target area (Thwaitessea-ice environment) and is able to generalize to different regions of Antarctica,reaching values of F1 = 80 %. A potential alternative version of the algorithm isproposed and discussed. This alternative score F1 values higher than F1 > 95 %for images of the target environment and F1 = 87 % for the image of the differentregion. However, it must not be confirmed as the final algorithm due to the needfor further verification. |
format |
Bachelor Thesis |
author |
Fuentes Soria, Carmen |
author_facet |
Fuentes Soria, Carmen |
author_sort |
Fuentes Soria, Carmen |
title |
Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region |
title_short |
Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region |
title_full |
Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region |
title_fullStr |
Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region |
title_full_unstemmed |
Creating a semantic segmentationmachine learning model for sea icedetection on radar images to study theThwaites region |
title_sort |
creating a semantic segmentationmachine learning model for sea icedetection on radar images to study thethwaites region |
publisher |
Luleå tekniska universitet, Rymdteknik |
publishDate |
2022 |
url |
http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-95352 |
long_lat |
ENVELOPE(-106.750,-106.750,-75.500,-75.500) |
geographic |
Thwaites Glacier |
geographic_facet |
Thwaites Glacier |
genre |
Antarc* Antarctica Sea ice Thwaites Glacier |
genre_facet |
Antarc* Antarctica Sea ice Thwaites Glacier |
op_relation |
http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-95352 |
op_rights |
info:eu-repo/semantics/openAccess |
_version_ |
1766273740387123200 |