Monogenic signal in seismic pattern recognition

Saavedra, Bernal Manzanilla (University of Alberta) | Sacchi, Mauricio (University of Alberta)


In seismic interpretation, the Hilbert transform has been a powerful aid for the interpreter. It allows representing the seismogram as a function of an instantaneous phase and amplitude envelope. This property enables estimating instantaneous temporal attributes such as an envelope, instantaneous phase, and instantaneous frequency. Unfortunately, the Hilbert transform applies only to one-dimensional signals. It is applied to individual seismic traces to generate spatio-temporal maps of seismic attributes. Alternatively, the Riesz transform can extract attributes directly from multidimensional signals. The Riesz transform is a generalization of the Hilbert transform to signals in Euclidean spaces of dimension greater than one. We discuss the application of the Riesz transform to generate the so-called monogenic signal which, in turn, is used to extract attributes from time slices of 3D seismic cubes.

Presentation Date: Tuesday, October 16, 2018

Start Time: 8:30:00 AM

Location: 210A (Anaheim Convention Center)

Presentation Type: Oral