SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images
Speckle reduction is a key step in many remote sensing applications. By strongly affecting synthetic aperture radar (SAR) images, it makes them difficult to analyze. Due to the difficulty to model the spatial correlation of speckle, a deep learning algorithm with semi-supervision is proposed in this...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9399231/ |
| _version_ | 1818620513500528640 |
|---|---|
| author | Emanuele Dalsasso Loic Denis Florence Tupin |
| author_facet | Emanuele Dalsasso Loic Denis Florence Tupin |
| author_sort | Emanuele Dalsasso |
| collection | DOAJ |
| description | Speckle reduction is a key step in many remote sensing applications. By strongly affecting synthetic aperture radar (SAR) images, it makes them difficult to analyze. Due to the difficulty to model the spatial correlation of speckle, a deep learning algorithm with semi-supervision is proposed in this article: SAR2SAR. Multitemporal time series are leveraged and the neural network learns to restore SAR images by only looking at noisy acquisitions. To this purpose, the recently proposed noise2noise framework <xref ref-type="bibr" rid="ref1">[1]</xref> has been employed. The strategy to adapt it to SAR despeckling is presented, based on a compensation of temporal changes and a loss function adapted to the statistics of speckle. A study with synthetic speckle noise is presented to compare the performances of the proposed method with other state-of-the-art filters. Then, results on real images are discussed, to show the potential of the proposed algorithm. The code is made available to allow testing and reproducible research in this field. |
| first_indexed | 2024-12-16T17:54:34Z |
| format | Article |
| id | doaj.art-f8fe2b8ab00344d19b5e2a854109c0d4 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-16T17:54:34Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-f8fe2b8ab00344d19b5e2a854109c0d42022-12-21T22:22:12ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01144321432910.1109/JSTARS.2021.30718649399231SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR ImagesEmanuele Dalsasso0https://orcid.org/0000-0001-7170-9015Loic Denis1https://orcid.org/0000-0001-9216-8318Florence Tupin2https://orcid.org/0000-0002-3110-8183LTCI, Télécom Paris, Institut Polytechnique de Paris, Palaiseau, FranceUJM-Saint-Etienne, CNRS, Institut d Optique Graduate School, Laboratoire Hubert Curien UMR 5516, Univ Lyon, Saint-Etienne, FranceLTCI, Télécom Paris, Institut Polytechnique de Paris, Palaiseau, FranceSpeckle reduction is a key step in many remote sensing applications. By strongly affecting synthetic aperture radar (SAR) images, it makes them difficult to analyze. Due to the difficulty to model the spatial correlation of speckle, a deep learning algorithm with semi-supervision is proposed in this article: SAR2SAR. Multitemporal time series are leveraged and the neural network learns to restore SAR images by only looking at noisy acquisitions. To this purpose, the recently proposed noise2noise framework <xref ref-type="bibr" rid="ref1">[1]</xref> has been employed. The strategy to adapt it to SAR despeckling is presented, based on a compensation of temporal changes and a loss function adapted to the statistics of speckle. A study with synthetic speckle noise is presented to compare the performances of the proposed method with other state-of-the-art filters. Then, results on real images are discussed, to show the potential of the proposed algorithm. The code is made available to allow testing and reproducible research in this field.https://ieeexplore.ieee.org/document/9399231/Deep learningimage despecklingsemi-supervisionsynthetic aperture radar (SAR) |
| spellingShingle | Emanuele Dalsasso Loic Denis Florence Tupin SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning image despeckling semi-supervision synthetic aperture radar (SAR) |
| title | SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images |
| title_full | SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images |
| title_fullStr | SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images |
| title_full_unstemmed | SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images |
| title_short | SAR2SAR: A Semi-Supervised Despeckling Algorithm for SAR Images |
| title_sort | sar2sar a semi supervised despeckling algorithm for sar images |
| topic | Deep learning image despeckling semi-supervision synthetic aperture radar (SAR) |
| url | https://ieeexplore.ieee.org/document/9399231/ |
| work_keys_str_mv | AT emanueledalsasso sar2sarasemisuperviseddespecklingalgorithmforsarimages AT loicdenis sar2sarasemisuperviseddespecklingalgorithmforsarimages AT florencetupin sar2sarasemisuperviseddespecklingalgorithmforsarimages |