dc.contributor.advisor | Doulgeris, Anthony Paul | |
dc.contributor.author | Lohse, Johannes | |
dc.date.accessioned | 2021-02-26T08:55:51Z | |
dc.date.available | 2021-02-26T08:55:51Z | |
dc.date.issued | 2021-03-12 | |
dc.description.abstract | With the Arctic sea ice continuously decreasing in both extent and thickness, fast and robust production of reliable ice charts becomes more important to ensure the safety of Arctic operations. This thesis focuses on the development of automated algorithms for the mapping of sea ice from synthetic aperture radar (SAR) images. It presents a thorough background on the topics of sea ice observations and ice charting, sea ice image classification, and the appearance of sea ice in SAR imagery. Three papers present the scientific developments in the thesis. Paper 1 focuses on the topic of feature selection. The study investigates the benefits of splitting a multi-class problem into several binary problems and selecting different feature sets specifically tailored towards these binary problems. Using a combination of classification accuracy and sequential search algorithms, the best order of classification steps and the optimal feature set for each class are found and combined into a numerically optimized decision tree. The method is tested on various examples, including an airborne, multi-frequency SAR data set over sea ice, and compared to traditional classification approaches. Paper 2 and 3 focus on the classification of Sentinel-1 (S1) wide-swath SAR images. Both papers use a newly generated training and validation data set for different sea ice types, which is is based on the visual analysis of overlapping S1 SAR and optical data. A particular challenge for the automated analysis of wide-swath SAR images is the surface-type dependent variation of backscatter intensity with incident angle (IA). In Paper 2, a novel method to directly incorporate this per-class IA effect into a classification algorithm is developed. Paper 3 investigates the IA dependence of texture features and extends the algorithm from Paper 2 to include textural information, in order to solve the ambiguities inherent in a classifier based on intensity only. | en_US |
dc.description.doctoraltype | ph.d. | en_US |
dc.description.popularabstract | Arctic sea ice has decreased in both thickness and extent over the past decades. Particularly, the older thicker multi-year ice has been disappearing, causing the Arctic sea ice cover to be on the verge of becoming seasonal. Besides implications for the Arctic ecosystems as well as the global weather and climate system, the continuous retreat of the sea ice cover makes the Arctic Ocean more easily accessible for human activity. Increasing Arctic operations require reliable sea ice observations over large areas and at high spatial and temporal resolution. Such monitoring is only possible with remote sensing methods from space. Synthetic aperture radar (SAR) can provide observations independently of sunlight or weather conditions, but interpretation and classification of sea ice in the images remain challenging.
In this thesis, new classification strategies are developed that address some of the main challenges in automated mapping of sea ice types in SAR imagery. Overlapping SAR and optical data are used to identify different ice types in the images and train the developed algorithms to automatically generate maps of sea ice type distribution. Integration with other data sources and observations will allow these methods to assist in the production of operational ice charts and continuously map the Arctic sea ice cover. This will contribute to ensuring the safety of Arctic operations, for the protection of both humans and the Arctic ecosystem.
In this thesis, new classification strategies are developed that address some of the main challenges in automated mapping of sea ice types in SAR imagery. Overlapping SAR and optical data are used to identify different ice types in the images and train the developed algorithms to automatically generate maps of sea ice type distribution. Integration with other data sources and observations will allow these methods to assist in the production of operational ice charts and continuously map the Arctic sea ice cover. This will contribute to ensuring the safety of Arctic operations, for the protection of both humans and the Arctic ecosystem. | en_US |
dc.description.sponsorship | The research in this thesis was funded by CIRFA partners and the Research Council of Norway (RCN) (grant number 237906) | en_US |
dc.identifier.uri | https://hdl.handle.net/10037/20606 | |
dc.language.iso | eng | en_US |
dc.publisher | UiT Norges arktiske universitet | en_US |
dc.publisher | UiT The Arctic University of Norway | en_US |
dc.relation.haspart | <p>Paper I: Lohse, J., Doulgeris, A.P. & Dierking, W. (2019). An Optimal Decision-Tree Design Strategy and Its Application to Sea Ice Classification from SAR Imagery. <i>Remote Sensing, 11</i>(13), 1574. Also available in Munin at <a href=https://hdl.handle.net/10037/17044>https://hdl.handle.net/10037/17044</a>.
<p>Paper II: Lohse, J., Doulgeris, A., & Dierking, W. (2020). Mapping sea-ice types from Sentinel-1 considering the surface-type dependent effect of incidence angle. Annals of Glaciology, 1-11. Also available in Munin at <a href=https://hdl.handle.net/10037/18738>https://hdl.handle.net/10037/18738</a>.
<p>Paper III: Lohse, J., Doulgeris, A.P. & Dierking, W. Incident Angle Dependence of Sentinel-1 Texture Features for Sea Ice Classification. (Submitted manuscript). Now published in <i>Remote Sensing, 2021, 13</i>(4), 552, available in Munin at <a href=https://hdl.handle.net/10037/20605>https://hdl.handle.net/10037/20605</a>. | en_US |
dc.relation.projectID | info:eu-repo/grantAgreement/RCN/SFI/237906/Norway/Centre for Integrated Remote Sensing and Forecasting for Arctic Operations/CIRFA/ | en_US |
dc.rights.accessRights | openAccess | en_US |
dc.rights.holder | Copyright 2021 The Author(s) | |
dc.subject.courseID | DOKTOR-004 | |
dc.subject | VDP::Mathematics and natural science: 400::Physics: 430::Electromagnetism, acoustics, optics: 434 | en_US |
dc.subject | VDP::Matematikk og Naturvitenskap: 400::Fysikk: 430::Elektromagnetisme, akustikk, optikk: 434 | en_US |
dc.subject | VDP::Mathematics and natural science: 400::Information and communication science: 420::Simulation, visualization, signal processing, image processing: 429 | en_US |
dc.subject | VDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420::Simulering, visualisering, signalbehandling, bildeanalyse: 429 | en_US |
dc.subject | VDP::Mathematics and natural science: 400::Mathematics: 410::Statistics: 412 | en_US |
dc.subject | VDP::Matematikk og Naturvitenskap: 400::Matematikk: 410::Statistikk: 412 | en_US |
dc.title | On Automated Classification of Sea Ice Types in SAR Imagery | en_US |
dc.type | Doctoral thesis | en_US |
dc.type | Doktorgradsavhandling | en_US |