Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images
Improving the geo-localization of optical satellite images is an important pre-processing step for many remote sensing tasks like monitoring by image time series or scene analysis after sudden events. These tasks require geo-referenced and precisely co-registered multi-sensor data. Images captured b...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2072-4292/9/6/586 |
| _version_ | 1818083098921795584 |
|---|---|
| author | Nina Merkle Wenjie Luo Stefan Auer Rupert Müller Raquel Urtasun |
| author_facet | Nina Merkle Wenjie Luo Stefan Auer Rupert Müller Raquel Urtasun |
| author_sort | Nina Merkle |
| collection | DOAJ |
| description | Improving the geo-localization of optical satellite images is an important pre-processing step for many remote sensing tasks like monitoring by image time series or scene analysis after sudden events. These tasks require geo-referenced and precisely co-registered multi-sensor data. Images captured by the high resolution synthetic aperture radar (SAR) satellite TerraSAR-X exhibit an absolute geo-location accuracy within a few decimeters. These images represent therefore a reliable source to improve the geo-location accuracy of optical images, which is in the order of tens of meters. In this paper, a deep learning-based approach for the geo-localization accuracy improvement of optical satellite images through SAR reference data is investigated. Image registration between SAR and optical images requires few, but accurate and reliable matching points. These are derived from a Siamese neural network. The network is trained using TerraSAR-X and PRISM image pairs covering greater urban areas spread over Europe, in order to learn the two-dimensional spatial shifts between optical and SAR image patches. Results confirm that accurate and reliable matching points can be generated with higher matching accuracy and precision with respect to state-of-the-art approaches. |
| first_indexed | 2024-12-10T19:32:36Z |
| format | Article |
| id | doaj.art-48c1671e811349f58ffa86e686c5392a |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-12-10T19:32:36Z |
| publishDate | 2017-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-48c1671e811349f58ffa86e686c5392a2022-12-22T01:36:12ZengMDPI AGRemote Sensing2072-42922017-06-019658610.3390/rs9060586rs9060586Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite ImagesNina Merkle0Wenjie Luo1Stefan Auer2Rupert Müller3Raquel Urtasun4German Aerospace Center (DLR), Remote Sensing Technology Institute, 82234 Wessling, GermanyDepartment of Computer Science University of Toronto, Toronto, ON M5S 3G, CanadaGerman Aerospace Center (DLR), Remote Sensing Technology Institute, 82234 Wessling, GermanyGerman Aerospace Center (DLR), Remote Sensing Technology Institute, 82234 Wessling, GermanyDepartment of Computer Science University of Toronto, Toronto, ON M5S 3G, CanadaImproving the geo-localization of optical satellite images is an important pre-processing step for many remote sensing tasks like monitoring by image time series or scene analysis after sudden events. These tasks require geo-referenced and precisely co-registered multi-sensor data. Images captured by the high resolution synthetic aperture radar (SAR) satellite TerraSAR-X exhibit an absolute geo-location accuracy within a few decimeters. These images represent therefore a reliable source to improve the geo-location accuracy of optical images, which is in the order of tens of meters. In this paper, a deep learning-based approach for the geo-localization accuracy improvement of optical satellite images through SAR reference data is investigated. Image registration between SAR and optical images requires few, but accurate and reliable matching points. These are derived from a Siamese neural network. The network is trained using TerraSAR-X and PRISM image pairs covering greater urban areas spread over Europe, in order to learn the two-dimensional spatial shifts between optical and SAR image patches. Results confirm that accurate and reliable matching points can be generated with higher matching accuracy and precision with respect to state-of-the-art approaches.http://www.mdpi.com/2072-4292/9/6/586geo-referencingmulti-sensor image matchingSiamese neural networksatellite imagessynthetic aperture radar |
| spellingShingle | Nina Merkle Wenjie Luo Stefan Auer Rupert Müller Raquel Urtasun Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images Remote Sensing geo-referencing multi-sensor image matching Siamese neural network satellite images synthetic aperture radar |
| title | Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images |
| title_full | Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images |
| title_fullStr | Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images |
| title_full_unstemmed | Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images |
| title_short | Exploiting Deep Matching and SAR Data for the Geo-Localization Accuracy Improvement of Optical Satellite Images |
| title_sort | exploiting deep matching and sar data for the geo localization accuracy improvement of optical satellite images |
| topic | geo-referencing multi-sensor image matching Siamese neural network satellite images synthetic aperture radar |
| url | http://www.mdpi.com/2072-4292/9/6/586 |
| work_keys_str_mv | AT ninamerkle exploitingdeepmatchingandsardataforthegeolocalizationaccuracyimprovementofopticalsatelliteimages AT wenjieluo exploitingdeepmatchingandsardataforthegeolocalizationaccuracyimprovementofopticalsatelliteimages AT stefanauer exploitingdeepmatchingandsardataforthegeolocalizationaccuracyimprovementofopticalsatelliteimages AT rupertmuller exploitingdeepmatchingandsardataforthegeolocalizationaccuracyimprovementofopticalsatelliteimages AT raquelurtasun exploitingdeepmatchingandsardataforthegeolocalizationaccuracyimprovementofopticalsatelliteimages |