High Resolution Radon Transforms based Compressive Sensing Principles
Y. Ma1 and Y. Luo1
Radon-based transform algorithms have been widely used in seismic data processing primarily for noise removal (surface waves and multiples). The basic assumption is sufficient dip or moveout difference between signal and noise. Additional sparseness criteria in the transform domain are useful constraints to minimize the overlapping of signal and noise. Theoretical studies based on compressive sensing principles have shown that under the sparse assumption, signals can be reconstructed from far less data or measurements than what is usually considered necessary according to the Nyquist sampling theory. This means that seismic signals in time-space domain could be represented and reconstructed from a few non-zero samples in the transform domain. The sparse representation of data in the transform domain offers opportunities to distinguish and suppress the unwanted noise in an efficient manner.
AAPG Search and Discovery Article #90188 ©GEO-2014, 11th Middle East Geosciences Conference and Exhibition, 10-12 March 2014, Manama, Bahrain