| 1. |
- Demchev, Denis, 1984, et al.
(författare)
-
Alignment of Multi-Frequency SAR Images Acquired over Sea Ice Using Drift Compensation
- 2023
-
Ingår i: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. - 2151-1535 .- 1939-1404. ; 16, s. 7393-7402
-
Tidskriftsartikel (refereegranskat)abstract
- We investigate the feasibility to align synthetic aperture radar (SAR) imagery based on a compensation for sea ice drift occurring between temporally shifted image acquisitions. The image alignment is a requirement for improving sea ice classification by combining multi-frequency SAR images acquired at different times. Images obtained at different radar frequencies provide complementary information, thus reducing ambiguities in the separation of ice types and the retrieval of sea ice parameters. For the alignment we use ice displacement vectors obtained from a sea ice drift retrieval algorithm based on pattern matching. The displacement vectors are organized on a triangular mesh and used for a piecewise affine transformation of the slave image onto the master image. In our case study we developed an alignment framework for pairs of ALOS-2 PALSAR-2 (L-band) and Sentinel-1 (C-band) images. We demonstrate several successful examples of alignment for time gaps ranging from a few hours to several days, depending on ice conditions. The data were acquired over three test sites in the Arctic: Belgica Bank, Fram Strait, and Lincoln Sea. We assess the quality of alignment using the structural similarity index (SSIM). From the displacement vectors, locations and extensions of patches of strong ice deformation are determined which allows to estimate the possible areal size of successful alignment over undeformed ice and a judgment of the expected quality for each image pair. The comprehensive assessment of hundreds of aligned L-C SAR pairs shows the potential of our method to work under various environmental conditions provided that the ice drift can be estimated reliably.
|
|
| 2. |
- Eriksson, Leif, 1970, et al.
(författare)
-
Alignment of L-and C-Band SAR Images for Enhanced Observations of Sea Ice
- 2022
-
Ingår i: International Geoscience and Remote Sensing Symposium (IGARSS). ; 2022-July, s. 3798-3801
-
Konferensbidrag (refereegranskat)abstract
- This paper discusses the alignment of Synthetic Aperture Radar (SAR) image pairs with one image acquired at L-and the other at C-band. For our study we used data from ALOS-2 PALSAR-2 and Sentinel-1 taken over sea ice in the Fram Strait. The use of multifrequency SAR data is beneficial for sea ice characterization and classification because it combines the advantages of the frequency dependence of signal penetration depths and backscattering characteristics due to small-scale ice properties. For drifting sea ice, however, an immediate combination of image pairs acquired with a time gap is usually not possible. Such cases require to consider the effects of ice drift. In this study we investigate the possibility to compensate for sea ice drift for separation in acquisition time ranging from a few hours to several days. The first results are promising and suggest that alignment of SAR image pairs obtained at different frequencies is possible even for larger time gaps between acquisitions if drift and deformation can be estimated reliably.
|
|
| 3. |
- Demchev, Denis, 1984, et al.
(författare)
-
Improving Sea Ice Drift Retrieval from SAR Images Using Phase- and Cross-Correlation Techniques
- 2020
-
Ingår i: Proceedings of the 2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering, EIConRus 2020. ; , s. 1378-1381
-
Konferensbidrag (refereegranskat)abstract
- A new combination of phase- and cross-correlation techniques for sea ice tracking from sequential synthetic aperture radar images investigated. An operational Python-based sea ice drift algorithm based on this combination from Sentinel-1 images is proposed.
|
|
| 4. |
- Demchev, Denis, 1984, et al.
(författare)
-
INTERCOMPARISON OF SEA ICE TRACKING ALGORITHMS FROM MULTIFREQUENCY SAR IMAGES IN THE ARCTIC
- 2023
-
Ingår i: IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM. - 2153-6996. - 9798350320107 ; , s. 158-160
-
Konferensbidrag (refereegranskat)abstract
- In this study, we assess the performance of the three state-of-the-art ice drift retrieval algorithms for Synthetic aperture radar (SAR) images. The algorithms were selected by their difference in underlying technique to reveal their advantages and limitations based on C- and X-band data in the Arctic. The results suggest that area-based algorithms are applicable in most cases, while feature-tracking algorithm demonstrates better precision but with fewer produced drift vectors. The hybrid algorithm provides the best data density compared to the other two algorithms but suffers from generating artificial vectors in the case of strong ice speed gradients such as at the edge between fast and drift ice.
|
|
| 5. |
- Demchev, Denis, 1984, et al.
(författare)
-
Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation
- 2023
-
Ingår i: Remote Sensing. - : MDPI AG. - 2072-4292. ; 15:5
-
Tidskriftsartikel (refereegranskat)abstract
- Permafrost tundra contains more than twice as much carbon as is currently in the atmosphere, and it is warming six times as fast as the global mean. Tundra lakes dynamics is a robust indicator of global climate processes, and is still not well understood. Satellite data, particularly, from synthetic aperture radar (SAR) is a suitable tool for tundra lakes recognition and monitoring of their changes. However, manual analysis of lake boundaries can be slow and inefficient; therefore, reliable automated algorithms are required. To address this issue, we propose a two-stage approach, comprising instance deep-learning-based segmentation by U-Net, followed by semantic segmentation based on a watershed algorithm for separating touching and overlapping lakes. Implementation of this concept is essential for accurate sizes and shapes estimation of an individual lake. Here, we evaluated the performance of the proposed approach on lakes, manually extracted from tens of C-band SAR images from Sentinel-1, which were collected in the Yamal Peninsula and Alaska areas in the summer months of 2015–2022. An accuracy of 0.73, in terms of the Jaccard similarity index, was achieved. The lake’s perimeter, area and fractal sizes were estimated, based on the algorithm framework output from hundreds of SAR images. It was recognized as lognormal distributed. The evaluation of the results indicated the efficiency of the proposed approach for accurate automatic estimation of tundra lake shapes and sizes, and its potential to be used for further studies on tundra lake dynamics, in the context of global climate change, aimed at revealing new factors that could cause the planet to warm or cool.
|
|
| 6. |
- Marbouti, Marjan, et al.
(författare)
-
Evaluating landfast sea ice ridging near UtqiagVik Alaska Using TanDEM-X interferometry
- 2020
-
Ingår i: Remote Sensing. - : MDPI AG. - 2072-4292. ; 12:8
-
Tidskriftsartikel (refereegranskat)abstract
- Seasonal landfast sea ice stretches along most Arctic coastlines and serves as a platform for community travel and subsistence, industry operations, and as a habitat for marine mammals. Landfast ice can feature smooth ice and areas of m-scale roughness in the form of pressure ridges. Such ridges can significantly hamper trafficability, but if grounded can also serve to stabilize the shoreward ice. We investigate the use of synthetic aperture radar interferometry (InSAR) to assess the formation and movement of ridges in the landfast sea ice near Utqiagvik, Alaska. The evaluation is based on the InSAR-derived surface elevation change between two TanDEM-X bistatic image pairs acquired during January 2012. We compare the results with backscatter intensity, coastal radar data, and SAR-derived ice drift and evaluate the utility of this approach and its relevance for evaluation of ridge properties, as well as landfast sea ice evolution, dynamics, and stability. © 2020 by the authors.
|
|
| 7. |
- Selyuzhenok, Valeria, et al.
(författare)
-
The bimodality of the East Siberian fast ice extent: mechanisms and changes
- 2023
-
Ingår i: Annals of Glaciology. - 0260-3055. ; In Press
-
Tidskriftsartikel (refereegranskat)abstract
- Using operational sea-ice maps, we provide first insight into the seasonal evolution of fast ice in the East Siberian Sea for the period between 1999 and 2021. The fast ice season tends to start later by 4.7 d per decade and to end earlier by 9.7 d per decade. As a result, there is a trend towards a shorter length of fast ice season by 2 weeks per decade. The analysis of air temperatures indicates that onset and end of the fast ice season are largely driven by thermodynamic processes. Two spatial modes (large, L-mode and small, S-mode) of East Siberian fast ice cover which have significant areal differences were distinguished. The occurrence of L- and S-modes was linked to the polarity of the Arctic Oscillation (AO) index. Negative AO phase leads to increased sea-ice convergence in the region, which in turn favours sea-ice grounding and promotes the development of large fast ice extent (L-mode). Lower deformation rates in the region during positive AO phase does not allow the formation of grounded features which results in small fast ice extent (S-mode). An analysis of sea-ice divergence confirms that L-mode seasons are characterised by higher on-shore convergence compared with S-mode seasons.
|
|
| 8. |
- Sudakow, Ivan, et al.
(författare)
-
MeltPondNet: A Swin Transformer U-Net for Detection of Melt Ponds on Arctic Sea Ice
- 2022
-
Ingår i: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. - 2151-1535 .- 1939-1404. ; 15, s. 8776-8784
-
Tidskriftsartikel (refereegranskat)abstract
- High-resolution aerial photographs of Arctic region are a great source for different sea ice feature recognition, which are crucial to validate, tune and improve climate models. Melt ponds on the surface of melting Arctic sea ice are of particular interest as they are sensitive and valuable indicators and are proxy to the processes in the Arctic climate system. Manual analysis of this remote sensing data is extremely difficult and time-consuming due to the complex shapes and unpredictable boundaries of the melt ponds, and that leads to the necessity for automatizing the processes. In this study, we propose a robust and efficient automatic method for melt pond region segmentation and boundary extraction from high-resolution aerial photographs. The proposed algorithm is based on a swin transformer U-Net in which we introduce novel cross-channel attention mechanisms into the decoder design. The framework operates with optical data and allows for classifying imagery into four classes: sea ice/snow, open water, melt pond, and submerged ice. We use aerial photographs collected during the Healy-Oden Trans Arctic Expedition (HO-TRAX) expedition over Arctic sea ice in the summer season of 2005 as test data. The experimental results show that the proposed method is suitable for precise automatic extraction of melt pond geometry and it can also be extended for other optical data sources that involve melt ponds. The approach has a promising potential to be used to analyze melt ponds' corresponding changes between years.
|
|
| 9. |
- Sultana, Aqsa, et al.
(författare)
-
R2UNet for melt pond detection
- 2023
-
Ingår i: Proceedings of SPIE - The International Society for Optical Engineering. - 0277-786X .- 1996-756X. ; 12527
-
Konferensbidrag (refereegranskat)abstract
- The massive shift in temperatures in the Arctic region has caused the increased Albedo effect as higher amount of solar energy is absorbed in the darker surface due to melting ice and snow. This continuous regional warming results in further melting of glaciers and loss of sea ice. Arctic melt ponds are important indicators of Arctic climate change. High-resolution aerial photographs are invaluable for identifying different sea ice features and are great source for validating, tuning, and improving climate models. Due to the complex shapes and unpredictable boundaries of melt ponds, it is extremely tedious, taxing, and time-consuming to manually analyze these remote sensing data that lead to the need for automatizing the technique. Deep learning is a powerful tool for semantic segmentation, and one of the most popular deep learning architectures for feature cascading and effective pixel classification is the UNet architecture. We introduce an automatic and robust technique to predict the bounding boxes for melt ponds using a Multiclass Recurrent Residual UNet (R2UNet) with UNet as a base model. R2UNet mainly consists of two important components in the architecture namely residual connection and recurrent block in each layer. The residual learning approach prevents vanishing gradients in deep networks by introducing shortcut connections, and the recurrent block, which provides a feedback connection in a loop, allows outputs of a layer to be influenced by subsequent inputs to the same layer. The algorithm is evaluated on Healy-Oden Trans Arctic Expedition (HO-TRAX) dataset containing melt ponds obtained during helicopter photography flights between 5 August and 30 September 2005. The testing and evaluation results show that R2UNet provides improved and superior performance when compared to UNet, Residual UNet (Res-UNet) and Recurrent U-Net (R-UNet).
|
|
