Remote sensing observations play an important role in the geo- and bioscientific community, since they enable various applications to accurately monitor the Earth and its changes – on a microscopic level as well as from space. Observation systems regularly provide data with a spectrally, spatially and temporally high resolution. Beside the challenge to deal with large amounts of data and limited class label information, current and future challenges comprise the definition and the way of integration of prior and domain knowledge, the learning of sophisticated features and the fusion of multiple sensor data.
My research aims at the development of machine learning methods, which are particularly designed for the analysis of 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, and scientific domain knowledge. I believe that it is important to develop methods which ensure a high discrimination power and at the same time model the underlying structure of the data. Moreover, to take the next step in understanding geo- and biophysical phenomena, both data science models and theory-based models need to be combined to form powerful theory-guided machine learning models.
– Ocean remote sensing and plant phenotyping
– (Deep) Representation/Feature learning
– Self-taught learning
– Sparse representation
– Incremental/Sequential learning
– Theory-guided machine learning