Refinement of Deconvolution by Neural Networks

  • Enders A. Robinson
Part of the Modern Approaches in Geophysics book series (MAGE, volume 21)


This chapter presents two seismic deconvolution applications of artificial neural networks. The first application involves a network based upon an adaptive linear combiner (ALC), in order to refine the results of deconvolution. For example, the conventional method of designing a shaping filter for a minimum-delay wavelet, requires the computation of a fixed filter by the method of least-squares. An alternative approach, considered in this chapter, involves an ALC incorporating a learning rule. A fixed filter vector, computed using least-squares, is used as the initial value of the weight vector for the ALC. The ALC then upgrades or refines this weight vector, in order to obtain the final refined filter coefficients. The second application uses a Hopfield neural network to correct for phase rotation, in order to produce an improved zero-phase seismic section, based upon well data.


Seismic Section Synthetic Seismogram Phase Rotation Hopfield Neural Network Seismic Trace 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Bozic, S. M., 1979, Digital and Kalman Filtering, Edward Arnold.Google Scholar
  2. Cioffi, J. M., and Byun, Y., 1993, Adaptive filtering, in Mitra, S. K., and Kaiser, J. F., Eds., Handbook for digital signal processing, John Wiley and Sons, New York, 1085–1142.Google Scholar
  3. Freeman, J. A., 1994, Simulating neural networks, Addison-Wesley, Reading, MA.Google Scholar
  4. Haykin, S. S., 1996, Adaptive filter theory, 3rd ed., Prentice-Hall, Upper Saddle River, NJ.Google Scholar
  5. Robinson, E. A., 1954, Predictive decomposition of time series with application to seismic exploration, Ph.D. thesis, M.I.T., Cambridge, MA, (reprinted in Geophysics, 32, 1967, 418–494.CrossRefGoogle Scholar
  6. Robinson, E. A., and Treitel, S., 1980, Geophysical signal analysis, Prentice-Hall, Englewood Cliffs.Google Scholar
  7. Robinson, E. A., and Durrani, T. S., 1986, Geophysical signal processing, Prentice-Hall International, UK Ltd, London.Google Scholar
  8. Robinson, E. A., and Osman, O. M., 1996, Deconvolution, Society of Exploration Geophysicists, Tulsa, Oklahoma.Google Scholar
  9. Wang, L., and Mendel, J. M., 1992, Adaptive minimum prediction-error deconvolution and source wavelet estimation using Hopfield neural networks, Geophysics, 57, 670–679.CrossRefGoogle Scholar
  10. Ziolkowski, A., 1984, Deconvolution, International Human Resources Dev. Corp., Boston, MA.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2003

Authors and Affiliations

  • Enders A. Robinson
    • 1
  1. 1.Krumb School of MinesColumbia UniversityNew YorkUSA

Personalised recommendations