The project aims at improving state-of-the-art oil slick detectors/classifiers by working out the following issues:
1. Combine ASAR and MERIS data to improve the classifier performance.
2. Adopt Bayesian region-based segmentation approaches using Markov Random Fields, leading to effective segmentation of dark regions in SAR images.
3. Use wind information derived from ASAR data and available on the Wave Mode Ocean Wave Spectra (ASA_WVP_2P).
4. Incorporate recent results, according to which high order moments (from the second on) of SAR images are informative with respect to oil/water classification.
5. Adopt Bayesian Networks to fuse ASAR and MERIS data and to build the classifier.
|Funding: FCT, ESA|
|Start Date: 01-12-2004|
|End Date: 01-12-2007|
|Team: José Manuel Bioucas Dias|
|Groups: Pattern and Image Analysis – Lx|
|Local Coordinator: José Manuel Bioucas Dias|