We have a particular focus on new Big-Data analytics methods for sub-surface data, such as sparse data representation and feature extraction of big sub-surface data. Sparse data representation aims to find a compact representation of data in the form of a linear combination of basis functions selected from a dictionary and allows, e.g., to represent seismic data in the wavelet domain with substantially reduced data sizes, to simplify the conditioning process. Also, we research the potential for automated feature extraction from seismic and rate data to aid model conditioning and decision making.
.png)