Group for Research in Decision Analysis

Unsupervised and self-taught learning for remote sensing image interpretation

Ribana Roscher University of Bonn, Germany

This presentation gives an overview about my current research with remote sensing application examples. In my research I am aiming at the development of pattern recognition methods, which are particularly designed for the analysis of large scale remote sensing data. I specifically focus on efficient classification methods, techniques for sophisticated feature learning and the integration of prior knowledge such as spatial and temporal information. A central idea in my research is to develop methods which ensure a high discrimination power and at the same time model the underlying structure of the data, since such methods are a prerequisite for the automatic analysis of earth observation data. More specifically, my main focus is on unsupervised and self-taught learning in order to integrate unlabeled data for the classifcation process. This covers at the moment mostly methods such as sparse representation, archetypal analysis and one-class classifier. My applications cover the analysis and interpretation of multi- and hyperspectral aerial and satellite images (LULC classification), but also the detection of unknown classes and anomaly detection.

Free entrance.
Welcome to everyone!