Knowledge extraction from Copernicus satellite data
Abstract We describe two alternative approaches of how to extract knowledge from high- and medium-resolution Synthetic Aperture Radar (SAR) images of the European Sentinel-1 satellites. To this end, we selected two basic types of images, namely images depicting arctic shipping routes with icebergs,...
| Published in: | IOP Conference Series: Earth and Environmental Science |
|---|---|
| Main Authors: | , , |
| Format: | Article in Journal/Newspaper |
| Language: | unknown |
| Published: |
IOP Publishing
2020
|
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1088/1755-1315/509/1/012014 https://iopscience.iop.org/article/10.1088/1755-1315/509/1/012014/pdf https://iopscience.iop.org/article/10.1088/1755-1315/509/1/012014 |
| Summary: | Abstract We describe two alternative approaches of how to extract knowledge from high- and medium-resolution Synthetic Aperture Radar (SAR) images of the European Sentinel-1 satellites. To this end, we selected two basic types of images, namely images depicting arctic shipping routes with icebergs, and - in contrast - coastal areas with various types of land use and human-made facilities. In both cases, the extracted knowledge is delivered as (semantic) categories ( i.e. , local content labels) of adjacent image patches from big SAR images. Then, machine learning strategies helped us design and validate two automated knowledge extraction systems that can be extended for the understanding of multispectral satellite images. |
|---|