Oil Spill Segmentation via Adversarial f-Divergence Learning
- Submitting institution
-
University of Exeter
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1817
- Type
- D - Journal article
- DOI
-
10.1109/TGRS.2018.2803038
- Title of journal
- IEEE Transactions on Geoscience and Remote Sensing
- Article number
- -
- First page
- 4973
- Volume
- 56
- Issue
- 9
- ISSN
- 0196-2892
- Open access status
- Compliant
- Month of publication
- February
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
-
- Citation count
- 13
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This interdisciplinary work develops an adversarial learning approach to address the challenges of segmenting irregular and noisy SAR images of oil spills, and achieves state of the art performance in real-world SAR images including Envisat, MODIS and NOWPAP. The proposed innovative model leads to subsequent wide adoptions in segmenting oil spill images (see: 10.1109/JSTARS.2018.2833485; 10.1109/JOE.2018.2842538; 10.1109/JSTARS.2020.2999961). The method supports the NERC project BigFoot (NE/P017436/1) to extract dynamic objects from visual data and a joint project with Plymouth Marine Lab funded by ESA Dragon Programme Phase 4, to detect early warning signals of water quality hazards from ESA Sentinel mission sensors.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -