Signal Analysis and Imaging Group

The Signal Analysis and Imaging Group carries out research in the area of seismic signal processing, inversion, and imaging.

Our research is funded by a group of industrial sponsors with interest in the development of advance techniques and algorithms for processing and imaging seismic data.

Research highlights

Sparse coding for random and coherent noise attenuation

(a) The basis patches found from sparse coding. (b) (left) Data from which the basis patches are found, (centre) ﬁltered data, and (right) the residuals.

a) Basis patches found by sparse coding. (b) The CMP gather from which the basis are found. (c) Multiples estimate found using sparse coding. (d) Multiple suppression using sparse coding. (e) Multiples estimate using the parabolic Radon transform. (f) Multiple suppression using the parabolic Radon transform. From Sparse-coding for data-driven coherent and incoherent noise attenuation by Kaplan, Sacchi and Ulrych, 2009 (submitted).

Interpolation with f-x adaptive filters

Adaptive interpolation of curved events via recursive prediction filters. a) Data prior to interpolation. b) Data after adaptive interpolation. From Naghizadeh and Sacchi, 2009, F-X Adaptive Seismic Trace Interpolation, Geophysics 72 (6), V111-V118.

Interpolation with local sparse Radon transforms

A sparse (high resolution) local Radon transform is used to reconstruct aliased data. From Sacchi et. al, 2007, Local Wavefield Operators, Radon Transforms and Sparsity, EAGE Workshop on Curvelets, London

Signal Analysis and Imaging Group
Department of Physics, University of Alberta
Edmonton, AB, T6G 2G7, Canada
Phone: (780) 492-5034 Fax: (780) 492-0714