Feature tracking for sea ice drift retrieval from SAR images
- Resource Type
- Conference
- Authors
- Demchev, Denis; Volkov, Vladimir; Kazakov, Eduard; Sandven, Stein
- Source
- 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) Geoscience and Remote Sensing Symposium (IGARSS), 2017 IEEE International. :330-333 Jul, 2017
- Subject
- Aerospace
Components, Circuits, Devices and Systems
Fields, Waves and Electromagnetics
Geoscience
Power, Energy and Industry Applications
Signal Processing and Analysis
Feature extraction
Radar tracking
Sea ice
Synthetic aperture radar
Kernel
feature tracking
sea ice drift
synthetic aperture radar
a-kaze
sentinel-1
- Language
- ISSN
- 2153-7003
A feature tracking techniques for sea ice drift retrieval from a pair of sequential satellite synthetic aperture radar (SAR) images are discussed. The Scale Invariant Feature Transform (SIFT), its alternative called ORB and A-KAZE features are selected for the intercomparison. The experimental results obtained for dual polarized Sentinel-1 C-SAR Extended Wide Swath mode data showed high relevance of feature tracking using nonlinear scale-spaces: A-KAZE. The approach exploits the benefits of nonlinear multi-scale image representations using A-KAZE features, which is a method that detects and describes image features in an anisotropic scale space that preserves important object boundaries while adaptively removing noise and small image details. It found that feature tracking using nonlinear scale-spaces on SAR images is preferable compared to other existing feature tracking alternatives that make use of Gaussian or linear scale spaces.