Automatic Selection of the Regularization Parameter in the Sparse Parabolic Radon Transform for Multiples Suppression

Authors

  • S.К. Karyakin PJSC Permneftegeofizika, 34 Lodygina Str., Perm 614090, Russia Author

DOI:

https://doi.org/10.17072/psu.geol.25.3.247

Keywords:

Radon transform, multiple waves, regularization, ℓ1-norm, L-curve, GCV, inverse problem, seismic processing, adaptive parameter selection

Abstract

The paper considers the problem of multiple suppression based on the high-resolution parabolic Radon transform, formulated as an ill-posed linear inverse problem. It is shown that ℓ1-regularization is natural for this formulation, since it agrees with the sparsity hypothesis for the model in the Radon domain and provides a higher selectivity of separation of primary and multiple waves compared with quadratic regularization. Central attention is paid to the choice of the regularization parameter λ, which simultaneously determines the stability of the solution, the degree of multiple suppression, and the preservation of the useful signal. A theoretically justified scheme for the objective choice of λ is proposed based on the L-curve method and the GCV-type criterion adapted to the nonlinear ℓ1 formulation through the effective number of freedom degrees. It is shown that for non-stationary seismic data the most promising strategy is not a global but a local windowed selection of the parameter λ. An algorithmic scheme for the implementation of the approach is presented, its limitations and directions for further testing on real data are discussed. The scientific novelty of the work consists in transferring objective criteria for choosing the regularization parameter to the formulation of the sparse Radon transform with local adaptation to the data structure. The feasibility of the approach is demonstrated by a numerical example on synthetic data reproducing the seismic and geological conditions of a platform section. The multiple energy is suppressed by 91 %, and the correlation coefficient between the recovered target reflection and the reference reaches 0.99.

Downloads

Download data is not yet available.

References

Тихонов А. Н. О решении некорректно поставленных задач и методе регуляризации // Доклады Академии наук СССР. 1963. Т. 151, № 3. С. 501–504.

Тихонов А. Н., Арсенин В. Я. Методы решения некорректных задач. М.: Наука, 1979.

Carmack P. S., Spence J. S., Schucany W. R. Generalised correlated cross-validation // Journal of Nonparametric Statistics. 2012. Vol. 24, No. 2.

P. 269–282.

Golub G. H., Heath M., Wahba G. Generalized cross-validation as a method for choosing a good ridge parameter // Technometrics. 1979. Vol. 21, No. 2. P. 215–223. DOI: 10.1080/00401706. 1979.10489751 EDN: HMRAZX

Hampson D. Inverse velocity stacking for multiple elimination // Journal of the Canadian Society of Exploration Geophysicists. 1986. Vol. 22, No. 1. P. 44–55.

Hansen P. C. Analysis of discrete ill-posed problems by means of the L-curve // SIAM Review. 1992. Vol. 34, No. 4. P. 561–580.

Hansen P. C., O'Leary D. P. The use of the L- curve in the regularization of discrete ill-posed problems // SIAM Journal on Scientific Computing. 1993. Vol. 14, No. 6. P. 1487–1503.

Hastie T., Tibshirani R., Friedman J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed. New York: Springer,

2009.

Park T., Casella G. The Bayesian Lasso // Journal of the American Statistical Association. 2008. Vol. 103, No. 482. P. 681–686.

Sacchi M. D., Ulrych T. J. High-resolution velocity gathers and offset space reconstruction // Geophysics. 1995. Vol. 60, No. 4. P. 1169–1177.

Stein C. M. Estimation of the mean of a multivariate normal distribution // The Annals of Statistics. 1981. Vol. 9, No. 6. P. 1135–1151.

Tibshirani R. Regression shrinkage and selection via the lasso // Journal of the Royal Statistical Society: Series B (Methodological). 1996. Vol. 58, No. 1. P. 267–288.

Trad D., Ulrych T., Sacchi M. Latest views of the sparse Radon transform // Geophysics. 2003. Vol. 68, No. 1. P. 386–399.

Wahba G. Spline Models for Observational Data. Philadelphia: SIAM, 1990.

Zhang Q., Wang H., Chen W., Huang G. A local Radon transform for seismic random noise attenuation // Journal of Applied Geophysics. 2021. Vol. 186. Art. 104264. DOI: 10.1016/j.jappgeo.2021. 104264 EDN: RVDOBX

Zou H., Hastie T., Tibshirani R. On the “degrees of freedom” of the lasso // The Annals of Statistics. 2007. Vol. 35, No. 5. P. 2173–2192.

Downloads

Published

2026-11-21

Issue

Section

GEOPHYSICS