Research
I'm interested in the application of machine learning to seismic data processing and migration,
3D surface wave inversion based on finitedifference method and
SPECFEM3D, seismic forward modeling
in frequency domain, superresolution imaging with surface waves, and natural migration of surface waves.
Representative papers are highlighted.


Multilayer Sparse Least Squares Migration= Deep Convolutional Neural Network
Zhaolun Liu and Gerard Schuster
arXiv:1904.09321, 2019
We recast the multilayered sparse inversion problem as a multilayered neural network problem.
Unlike standard least squares migration (LSM) which finds the optimal reflectivity image,
neural network least squares migration (NNLSM) finds both the optimal reflectivity image and the quasimigrationGreen's functions.
These quasimigrationGreen's functions are also denoted as the convolutional filters in a convolutional neural network (CNN) and are similar to
migration Green's functions. We show that the CNN filters and feature maps are directly related to the migration Green's functions and reflectivity distributions.
Thus, we provide for the first time a physical interpretation of the filters and feature maps in deep CNN in terms of the operators for seismic imaging.
The advantage of NNLSM over standard LSM is that its computational cost is significantly less and it can be used for denoising coherent and incoherent
noise in migration images. Its disadvantage is that the NNLSM reflectivity image is only an approximation to the actual reflectivity distribution.
However, the approximate reflectivity image can be used as a superresolution attribute image for highresolution delineation of geologic bodies.


3D Waveequation Dispersion Inversion for Data Recorded on Irregular Topography
Zhaolun Liu
CSIM Annual Report, 2019
SEG Annual Meeting, 2019
Irregular topography can cause strong scattering and defocusing of propagating surface
waves. Thus it is important to consider such effects when inverting surface waves for the
shallow Svelocity structures. Here, we present a 3D surfacewave dispersion inversion
method that takes into account the topographic effects modeled by a 3D spectral element
solver.


Convolutional Sparse Coding for Noise Attenuation of Seismic Data
Zhaolun Liu,
Kai Lu and Xiaodan Ge
SEG Maximizing Asset Value through Artificial Intelligence and Machine Learning Workshop, 2018
The seismic data with a relatively high signaltonoise ratio are chosen for training to get the learned basis functions. Then we use all (or a subset) of the basis functions to attenuate the random or coherent noise in
the seismic data.


Neural Network Least Squares Migration
Zhaolun Liu,
and Gerard Schuster
First EAGE/SBGf Worskhop on Least Squares Migration, Rio de Janeiro, Brazil, 2018
EAGE Annual Meeting, 2019
Sparse least squares migration (SLSM) estimates the reflectivity distribution
that honors a sparsity condition. This problem can be reformulated by finding both the
sparse coefficients and basis functions from the data to
predict the migration image. This is designated as neural network least squares migration (NLSM),
which is a more general formulation of SLSM.


Multiscale and LayerStripping WaveEquation Dispersion Inversion of Rayleigh Waves
Zhaolun Liu and
Lianjie Huang
GJI, 2019
The multiscale and layerstripping method can alleviate the local minimum problem of waveequation dispersion inversion of Rayleigh waves.


3D Waveequation Dispersion Inversion of Rayleigh Waves
Zhaolun Liu, Jing Li,
Sherif Hanafy, and
Gerard Schuster
Geophyscis, 2019
We extend the 2D waveequation dispersion inversion (WD) method to 3D waveequation inversion of surface waves for the shearvelocity distribution.


Semistationary Supervirtual Interferometry of Reflections and Diving Waves
Kai Lu, Zhaolun Liu, and Xiaodan Ge
CSIM Annual Report, 2018
we extend the application of SVI to faroffset reflections and diving waves by defining semistationary phases.
Semistationary phases mean that the phase difference between adjacent traces in the common pair gather
(CPG) are very small, so that stacking the semistationary traces with techniques of limiting the stacking zone and phase shift compensation also enhances the SNR.


Imaging nearsurface heterogeneities by natural migration of backscattered surface waves: Field data test
Zhaolun Liu,
Abdullah AlTheyab,
Sherif Hanafy, and
Gerard Schuster
Geophyscis, 2017
SEG_PPT,
bibtex
We have developed a methodology for detecting the pres
ence of nearsurface heterogeneities by naturally migrating
backscattered surface waves in controlledsource data.
This natural migration method does not require knowledge of the nearsurface phasevelocity
distribution because it uses the recorded data to approximate the Green’s functions for migration.


Superresolution nearfield imaging with surface waves
Lei Fu, Zhaolun Liu and
Gerard Schuster
Geophys. J. Int, 2017
We present the theory for nearfield superresolution imaging with surface waves and time reverse mirrors (TRMs).


An optimized implicit finitedifference scheme for the twodimensional Helmholtz equation
Zhaolun Liu, Peng Song, Jinshan Li and Xiaobo Zhang
Geophys. J. Int, 2015
PPT
We have developed an implicit finitedifference scheme for the 2D Helmholtz equation.

