Deep Learning for Seismic Inverse Problems: Toward the Acceleration of Geophysical Analysis Workflows

被引：82

作者：

Adler, Amir ^{[1
,2
,3
,4
,5
,6
]}

Araya-Polo, Mauricio ^{[7
,8
,9
,10
,11
]}

Poggio, Tomaso ^{[2
,12
,13
,14
,15
,16
]}

机构：

[1] Braude Coll Engn, Elect Engn Dept, IL-2161002 Karmiel, Israel

[2] MIT, McGovern Inst Brain Res, 77 Massachusetts Ave, Cambridge, MA 02139 USA

[3] MIT, Ctr Brains Minds & Machines, 77 Massachusetts Ave, Cambridge, MA 02139 USA

[4] MIT, Deep Learning Solut, Royal Dutch Shell, 77 Massachusetts Ave, Cambridge, MA 02139 USA

[5] Minist Commun, Jerusalem, Israel

[6] NextWave Wireless Inc, Wi Fi Div, AT&T, San Diego, CA USA

[7] Total E&P RT, Houston, TX 77002 USA

[8] Rice Univ, Computat & Appl Math Dept, Houston, TX USA

[9] Shell Int E&P Inc, Geophys & Machine Learning Efforts, Tananger, Sola, Norway

[10] Repsol USA, Warrendale, PA USA

[11] Barcelona Supercomp Ctr, Kaleidoscope Project, Barcelona, Spain

[12] MIT, Dept Brain & Cognit Sci, E25-618, Cambridge, MA 02139 USA

[13] MIT, Comp Sci & Artificial Inelligence Lab, 77 Massachusetts Ave, Cambridge, MA 02139 USA

[14] MIT, Ctr Biol & Computat Learning, 77 Massachusetts Ave, Cambridge, MA 02139 USA

[15] McGovern Inst, Ctr Brains Minds & Machines, Cambridge, MA USA

[16] MIT Quest Intelligence, Cambridge, MA USA

来源：

IEEE SIGNAL PROCESSING MAGAZINE | 2021年 / 38卷 / 02期

关键词：

Seismic measurements; Inverse problems; Earthquakes; Deep learning; Hydrocarbons; Hazards; Earth; Analytical models; Geophysical measurements; GENERATIVE ADVERSARIAL NETWORKS; WAVE-FORM INVERSION;

D O I：

10.1109/MSP.2020.3037429

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Seismic inversion is a fundamental tool in geophysical analysis, providing a window into Earth. In particular, it enables the reconstruction of large-scale subsurface Earth models for hydrocarbon exploration, mining, earthquake analysis, shallow hazard assessment, and other geophysical tasks.

引用

页码：89 / 119

页数：31

共 66 条

[1] Adler A., 2019, 81 EAGE C EXH 2019, V2019, P1, DOI [DOI 10.3997/2214-4609.201901512, 10.3997/2214-4609.201901512]
[2] Alfarraj M., 2019, SEG TECHNICAL PROGRA, P2298, DOI DOI 10.1190/SEGAM2019-3215902.1
[3] Alfarraj M., 2018, SEG INT EXP 88 ANN M, P2141, DOI DOI 10.1190/SEGAM2018-2995752.1
[4] [Anonymous], 2018, ARXIV180107232
[5] [Anonymous], 2018, SEG TECHNICAL PROGRA, DOI [DOI 10.1190/SEGAM2018-2998603.1, 10.1190/segam2018-2998603.1]
[6] Deep learning-driven velocity model building workflow
Araya-Polo M.
Farris S.
Florez M.
[J]. Leading Edge, 2019, 38 (11) : 872A1 - 872A9
[7] Araya-Polo Mauricio, 2018, Leading Edge, V37, P58, DOI 10.1190/tle37010058.1
[8] Assessing Accelerator-Based HPC Reverse Time Migration
Araya-Polo, Mauricio
Cabezas, Javier
Hanzich, Mauricio
Pericas, Miquel
Rubio, Felix
Gelado, Isaac
Shafiq, Muhammad
Morancho, Enric
Navarro, Nacho
Ayguade, Eduard
Maria Cela, Jose
Valero, Mateo
[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2011, 22 (01) : 147 - 162
[9] ACOUSTIC IMPEDANCE LOGS COMPUTED FROM SEISMIC TRACES
BECQUEY, M
LAVERGNE, M
WILLM, C
[J]. GEOPHYSICS, 1979, 44 (09) : 1485 - 1501
[10] Representation Learning: A Review and New Perspectives
Bengio, Yoshua
Courville, Aaron
Vincent, Pascal
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2013, 35 (08) : 1798 - 1828

← 1 2 3 4 5 6 7 →