Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway
Remote sensing is a key tool to derive glacier surface velocities but existing mapping methods, such as cross-correlation techniques, can fail where surface properties change temporally or where large velocity variations occur spatially. High-resolution datasets, such as UAV imagery, offer a promisi...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-05-01
|
| Series: | Frontiers in Remote Sensing |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/frsen.2025.1586933/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849325468699852800 |
|---|---|
| author | Harald Zandler Jakob Abermann Benjamin A. Robson Alexander Maschler Thomas Scheiber Jonathan L. Carrivick Jacob C. Yde |
| author_facet | Harald Zandler Jakob Abermann Benjamin A. Robson Alexander Maschler Thomas Scheiber Jonathan L. Carrivick Jacob C. Yde |
| author_sort | Harald Zandler |
| collection | DOAJ |
| description | Remote sensing is a key tool to derive glacier surface velocities but existing mapping methods, such as cross-correlation techniques, can fail where surface properties change temporally or where large velocity variations occur spatially. High-resolution datasets, such as UAV imagery, offer a promising solution to tackle these issues and to study small-scale glacier dynamics, but new workflows are required to handle such data. Therefore, we tested the potential of new deep learning-based image-matching algorithms for deriving glacier surface velocities across the ablation area of a glacier with strong spatial variability in surface velocities (<5 m/yr to >100 m/yr) and substantial changes in surface properties between image acquisitions. For a thorough comparison of state-of-the-art methods and sensors, we applied three different techniques (cross-correlation using geoCosiCorr3D, feature tracking with ORB using SeaIceDrift and the new deep learning-based method using ICEpy4D) and three different platforms (Sentinel-2, PlanetScope, UAVs) to estimate glacier surface velocities. Results showed lowest errors for velocities derived with the deep learning-based approach applied to UAV imagery (RMSE = 2.17 m/yr, R2 = 0.99), followed by cross-correlation using Sentinel-2 images (RMSE = 21.0 m/yr, R2 = 0.59) and the deep learning-based approach with PlanetScope data (RMSE = 21.28 m/yr, R2 = 0.36). Cross-correlation with geoCosiCorr3D resulted in comparably high errors with the UAV dataset (RMSE = 36.22 m/yr, R2 = 0.24), whereas ORB-based feature tacking showed lowest performance with all sensors. Spatial patterns of computed velocities indicate that applying existing cross-correlation methods for areas with regular displacements or low glacier velocities yields suitable results on UAV data, but innovative deep learning-based approaches are required for resolving rapid changes in velocities or in surface properties. This novel method benefits from improved keypoint detection and matching through training using neural networks and data characterized by challenging geometries, outlier minimization and more robust descriptors by applying cross-attention layers. We conclude that continued development of deep learning-based feature tracking approaches for glacier velocity computations may substantially improve UAV-based velocity derivations applied to challenging situations. This method is able to deliver reliable displacement data in situations where traditional methods fail, which implies a new level of detail in understanding and interpreting glacier dynamics. |
| format | Article |
| id | doaj-art-1b28062f8e3341b08fcc0f677c274411 |
| institution | Kabale University |
| issn | 2673-6187 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Remote Sensing |
| spelling | doaj-art-1b28062f8e3341b08fcc0f677c2744112025-08-20T03:48:23ZengFrontiers Media S.A.Frontiers in Remote Sensing2673-61872025-05-01610.3389/frsen.2025.15869331586933Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, NorwayHarald Zandler0Jakob Abermann1Benjamin A. Robson2Alexander Maschler3Thomas Scheiber4Jonathan L. Carrivick5Jacob C. Yde6Department of Geography and Regional Science, University of Graz, Graz, AustriaDepartment of Geography and Regional Science, University of Graz, Graz, AustriaDepartment of Earth Science, University of Bergen, Bergen, NorwayDepartment of Civil Engineering and Environmental Sciences, Western Norway University of Applied Sciences, Sogndal, NorwayDepartment of Civil Engineering and Environmental Sciences, Western Norway University of Applied Sciences, Sogndal, NorwaySchool of Geography and Water@leeds, University of Leeds, Leeds, United KingdomDepartment of Civil Engineering and Environmental Sciences, Western Norway University of Applied Sciences, Sogndal, NorwayRemote sensing is a key tool to derive glacier surface velocities but existing mapping methods, such as cross-correlation techniques, can fail where surface properties change temporally or where large velocity variations occur spatially. High-resolution datasets, such as UAV imagery, offer a promising solution to tackle these issues and to study small-scale glacier dynamics, but new workflows are required to handle such data. Therefore, we tested the potential of new deep learning-based image-matching algorithms for deriving glacier surface velocities across the ablation area of a glacier with strong spatial variability in surface velocities (<5 m/yr to >100 m/yr) and substantial changes in surface properties between image acquisitions. For a thorough comparison of state-of-the-art methods and sensors, we applied three different techniques (cross-correlation using geoCosiCorr3D, feature tracking with ORB using SeaIceDrift and the new deep learning-based method using ICEpy4D) and three different platforms (Sentinel-2, PlanetScope, UAVs) to estimate glacier surface velocities. Results showed lowest errors for velocities derived with the deep learning-based approach applied to UAV imagery (RMSE = 2.17 m/yr, R2 = 0.99), followed by cross-correlation using Sentinel-2 images (RMSE = 21.0 m/yr, R2 = 0.59) and the deep learning-based approach with PlanetScope data (RMSE = 21.28 m/yr, R2 = 0.36). Cross-correlation with geoCosiCorr3D resulted in comparably high errors with the UAV dataset (RMSE = 36.22 m/yr, R2 = 0.24), whereas ORB-based feature tacking showed lowest performance with all sensors. Spatial patterns of computed velocities indicate that applying existing cross-correlation methods for areas with regular displacements or low glacier velocities yields suitable results on UAV data, but innovative deep learning-based approaches are required for resolving rapid changes in velocities or in surface properties. This novel method benefits from improved keypoint detection and matching through training using neural networks and data characterized by challenging geometries, outlier minimization and more robust descriptors by applying cross-attention layers. We conclude that continued development of deep learning-based feature tracking approaches for glacier velocity computations may substantially improve UAV-based velocity derivations applied to challenging situations. This method is able to deliver reliable displacement data in situations where traditional methods fail, which implies a new level of detail in understanding and interpreting glacier dynamics.https://www.frontiersin.org/articles/10.3389/frsen.2025.1586933/fullUAVPlanetScopeSentinel-2cross-correlationSuperpointSuperGlue |
| spellingShingle | Harald Zandler Jakob Abermann Benjamin A. Robson Alexander Maschler Thomas Scheiber Jonathan L. Carrivick Jacob C. Yde Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway Frontiers in Remote Sensing UAV PlanetScope Sentinel-2 cross-correlation Superpoint SuperGlue |
| title | Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway |
| title_full | Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway |
| title_fullStr | Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway |
| title_full_unstemmed | Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway |
| title_short | Deep learning outperforms existing algorithms in glacier surface velocity estimation with high-resolution data – the example of Austerdalsbreen, Norway |
| title_sort | deep learning outperforms existing algorithms in glacier surface velocity estimation with high resolution data the example of austerdalsbreen norway |
| topic | UAV PlanetScope Sentinel-2 cross-correlation Superpoint SuperGlue |
| url | https://www.frontiersin.org/articles/10.3389/frsen.2025.1586933/full |
| work_keys_str_mv | AT haraldzandler deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway AT jakobabermann deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway AT benjaminarobson deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway AT alexandermaschler deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway AT thomasscheiber deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway AT jonathanlcarrivick deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway AT jacobcyde deeplearningoutperformsexistingalgorithmsinglaciersurfacevelocityestimationwithhighresolutiondatatheexampleofausterdalsbreennorway |