Multiscale and supervised image analysis in earth observation and remote sensing – University of Copenhagen

Multiscale and supervised image analysis in earth observation and remote sensing

Talk By Sébastien Lefèvre; Full Professor in Computer Science, Univ. Bretagne-Sud / IRISA

Abstract:

Observation is one of the key issues in the understanding of environmental systems. Remote sensing provides an invaluable source of data in this context, but it requires some specific techniques to be developed to face the intrinsic complexity within the data, i.e. a massive amount of multidimensional (multi- or hyperspectral) noisy observations with high spatiotemporal variability. In this talk, I will review some of the research activities recently conducted in the OBELIX team from IRISA research institute (www.irisa.fr/obelix<http://www.irisa.fr/obelix>) in this context. More precisely, I will present how new developments in machine learning and multiscale image analysis can help to improve the processing of remote sensing data for earth observation.

Bio:

Sébastien Lefèvre received his M.Sc. and Eng. degrees from the University of Technology of Compiègne in 1999, and his Ph.D. from the University of Tours in 2002, and his habilitation in 2009 from the University of Strasbourg. From 2003 to 2010, he was an Associate Professor in the Department of Computer Sciences and the Image Sciences at University of Strasbourg. In 2010, he joined the University of Bretagne-Sud as a Full Professor in Computer Science, in the Institute of Technology of Vannes and in IRISA where he is leading the OBELIX team dedicated to image analysis and machine learning for remote sensing and earth observation (www.irisa.fr/obelix). He has coauthored more than 100 papers in image analysis and pattern recognition. His current research interests are mainly related to multivariate mathematical morphology, hierarchical models and machine learning applied to remote sensing of environment.