Blind source separation methods and application to processing remote sensing data
Webinar Speaker: Moussa Sofiane Karoui, Agence Spatiale Algérienne, Centre des Techniques Spatiales, Arzew, Algeria
- Friday, 28 May 2021
- 4pm Central European Time
- Sponsored by GRSS
About the Webinar
In this talk, firstly, the problem known as blind source separation (BSS) as well as the main associated methods will be considered: a general overview of this problem as well as the main classes of methods will be presented. Then, and in the field of remote sensing, unsupervised unmixing methods, which are related to the BSS problem, will be considered. These methods consist in estimating spectra (with/without spectral variability) of materials present in an imaged scene as well as their abundance fractions. Particular attention will be paid to constrained matrix factorization unmixing-based methods. Finally, other methods, allowing some unmixing-based functionalities/applications in the remote sensing image processing field, such as sharpening, object/change detection and shadow compensation, will also be exposed.
Speakers’ Bio
Moussa Sofiane Karoui was born in Oran, Algeria, in 1979. In 2000, he graduated in Applied Mathematics from the Université Es-Sénia, Oran, Algeria. In 2007, he received his M.Sc. degree in Signal and Image Processing from the Université Paul Sabatier Toulouse 3, Toulouse, France. In 2008, he received his Magister’s degree in Signal and Image Processing from the Université des Sciences et de la Technologie, Oran, Algeria. In 2012, he received his Ph.D. degree in Signal and Image Processing from the Université Paul Sabatier Toulouse 3, Toulouse, France, co-supervised with the Université des Sciences et de la Technologie, Oran, Algérie. Since 2001, he is a Senior Researcher at the Centre des Techniques Spatiales (Agence Spatiale Algérienne), Arzew, Algeria. His current major activities include linear/nonlinear spectral/spatial unmixing techniques in remote sensing imagery, statistical and especially, BSS methods with applications to remote sensing imagery.