Non-negative Matrix Factorization with Sparseness Constraints for Sea Ice SAR Feature Extraction

Conference: EUSAR 2008 - 7th European Conference on Synthetic Aperture Radar
06/02/2008 - 06/05/2008 at Friedrichshafen, Germany

Proceedings: EUSAR 2008

Pages: 4Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Karvonen, Juha (Finnish Institute of Marine Research (FIMR), PB 2, FIN-00561, Helsinki, Finland)

In this paper the Non-negative Matrix Factorization (NMF) with sparseness constraints is studied for sea ice SAR feature extraction. NMF can, among other things, be used to extract certain basis vectors (elementary features) from image data. The sparseness constrained NMF (SC-NMF) can be used to restrict the sparseness of the decomposition of data, and thus it makes possible to select sets of features different in their nature. These features can then be used in classifying SAR texture of sea ice SAR images.