IEEE Signal Processing - March 2018 - 96

Attention models inspired by the HVS

We expect to
witness more
breakthroughs in the
near future in seismic
interpretation-enabled
subsurface understanding.

In addition to machine learning, we believe the
other important factor that will impact seismic
interpretation research is the HVS modeling.
For visual analysis, HVS reduces the amount
of the sensory data information, also known as
the task-free visual search, by focusing on the
perceptually salient segments of visual data that
conveys the most useful information about the
scene. For example, as an interpreter traverses through a seismic
volume or a video, the HVS that is sensitive to structures and
variations in amplitude and surrounding environment registers
the relative motion of seismic structures across all three dimensions, not all of which are characterized by strong seismic reflections or texture contrast. Therefore, computational algorithms
inspired by HVS can be used to develop attention models that
can effectively automate the structural interpretation of seismic
volumes. Among many, one such attention model is defined
as a saliency model, which attempts to predict the interesting areas in images and videos, typically called salient regions,
by relying on low-level visual cues. In seismic interpretation,
visual saliency is important to predict the attention of human
interpreters and highlight areas of interest in seismic sections,
which cannot only help in the automation of the interpretation
process but also develop attention models that can be applied
to new data sets. Recently, a seismic attribute called SalSi was
proposed in [21] and [82], which highlights salient areas within
seismic volumes to assist interpreters in structural interpretation of salt domes and faults. Furthermore, by incorporating a
priori information into saliency detection, attention models can
be used to mimic the behavior of interpreters looking at seismic
sections. Therefore, saliency detection is one of the promising
directions that can not only assist interpreters in the interpretation process but also help develop better attention models for
seismic interpretation.
It is a challenging task to train a machine to understand
and reveal various structures of Earth's subsurface. Yet, once
achieved, it will be a truly rewarding accomplishment that
has a large impact on the economy and society. Through its
fast advancement in the past decade, seismic interpretation
has already demonstrated its high potential in fulfilling this
mission. With a broader and deeper collaboration between the
geophysical and the signal processing communities, we expect
to witness more breakthroughs in the near future in seismicinterpretation-enabled subsurface understanding.

Acknowledgments
We would like to acknowledge the support of the Center for Energy and Geo Processing at the Georgia Institute of Technology,
and King Fahd University of Petroleum and Minerals (Project
GTEC 1401-1402).

Authors
Ghassan AlRegib (alregib@gatech.edu) received his B.S degree
in electrical engineering in 1997 from King Fahd University of
Petroleum and Minerals, Dhahran, Saudi Arabia, and his Ph.D.
96

degree in electrical and computer engineering in 2003 from the Georgia Institute
of Technology (Georgia Tech), where he is
a professor in the School of Electrical and
Computer Engineering. In 2012, he was
named the director of Georgia Tech's
Center for Energy and Geo Processing. He
has authored/coauthored more than 200
articles in international journals and conference proceedings and has been issued several U.S. patents. He
is a member of the IEEE Image, Video, and Multidimensional
Signal Processing Workshop and the IEEE International
Workshop on Multimedia Signal Processing Technical
Committees (2015-2020). He received several awards including
the ECE Outstanding Junior Faculty Member Award and the
Steven A. Denning Faculty Award for Global Engagement. He
has been an active consultant to global corporations and educational organizations as well as a witness expert on several patent
infringement cases. He is the technical program chair of the
2020 IEEE International Conference on Image Processing. He is
a Senior Member of the IEEE.
Mohamed Deriche (mderiche@kfupm.edu.sa) received his
B.Sc. degree in electrical engineering from the National
Polytechnic School, Algeria, in 1986, and his Ph.D. degree in
signal processing from the University of Minnesota in 1994.
He worked at Queensland University of Technology, Australia,
before joining King Fahd University of Petroleum and Minerals
(KFUPM) in Dhahran, Saudi Arabia, where he leads the signal
processing group. He has published more than 200 papers in
multimedia signal and image processing. He delivered numerous invited talks and chaired several conferences including
GlobalSIP-MPSP, IEEE Gulf (GCC), Image Processing Tools
and Applications, and TENCON (a Region 10 conference).
He has supervised more than 35 M.Sc. and Ph.D. students
and is the recipient of the IEEE Third Millennium Medal. He
also received the Shauman Best Researcher Award, and both
the Excellence in Research and Excellence in Teaching
Awards at KFUPM.
Zhiling Long (zhiling.long@ece.gatech.edu) received his
Ph.D. degree in electrical engineering from Mississippi State
University. He is currently a postdoctoral fellow in the Center
for Energy and Geo Processing at the Georgia Institute of Technology. His research interests include signal and image processing, machine learning, and human perception models, with
applications to seismic interpretation and computer vision.
Haibin Di (haibin.di@ece.gatech.edu) received his B.E.
degree in geophysics from Ocean University of China, in
2008 and his Ph.D. degree in geology from West Virginia
University in 2016. In summer 2016, he joined the Center for
Energy and Geo Processing at the Georgia Institute of
Technology as a postdoctoral fellow. His primary research
interests include machine-learning-assisted three-dimensional
seismic interpretation, reservoir characterization, and seismic
attribute analysis.
Zhen Wang (zwang313@gatech.edu) received his B.Sc. and
M.S. degrees in information and communication engineering

IEEE Signal Processing Magazine

March 2018

Table of Contents for the Digital Edition of IEEE Signal Processing - March 2018