Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO

Y. Cui; O. R. Huff; U. B. Waheed; J. E. Lie; A. K. Evensen; A. J. Bugge

doi:10.3997/2214-4609.202510766

Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO

Y. Cui
, O. R. Huff
, U. B. Waheed
, J. E. Lie
, A. K. Evensen
, A. J. Bugge

Department of Geosciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This study aims to develop and validate a novel neural network architecture, KFNO, for reconstructing missing near-offset traces in marine seismic data. By integrating the Fourier Neural Operator (FNO) and the Kolmogorov-Arnold Network (KAN), the KFNO model aims to address the positional disparities between source and receiver lines that lead to missing traces, enhancing the accuracy and reliability of seismic data processing and interpretation.

Original language	English
Title of host publication	86th EAGE Annual Conference and Exhibition
Publisher	European Association of Geoscientists and Engineers, EAGE
ISBN (Electronic)	9789462825352
DOIs	https://doi.org/10.3997/2214-4609.202510766
State	Published - 2025
Event	86th EAGE Annual Conference and Exhibition - Toulouse, France Duration: 2 Jun 2025 → 5 Jun 2025

Publication series

Name	86th EAGE Annual Conference and Exhibition

Conference

Conference	86th EAGE Annual Conference and Exhibition
Country/Territory	France
City	Toulouse
Period	2/06/25 → 5/06/25

Bibliographical note

Publisher Copyright:
© 2025 86th EAGE Annual Conference and Exhibition. All rights reserved.

ASJC Scopus subject areas

Geochemistry and Petrology
Geophysics

Access to Document

10.3997/2214-4609.202510766

Cite this

@inproceedings{4158f7b8c9cd4ac2b3fc6a905daab7ef,

title = "Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO",

abstract = "This study aims to develop and validate a novel neural network architecture, KFNO, for reconstructing missing near-offset traces in marine seismic data. By integrating the Fourier Neural Operator (FNO) and the Kolmogorov-Arnold Network (KAN), the KFNO model aims to address the positional disparities between source and receiver lines that lead to missing traces, enhancing the accuracy and reliability of seismic data processing and interpretation.",

author = "Y. Cui and Huff, \{O. R.\} and Waheed, \{U. B.\} and Lie, \{J. E.\} and Evensen, \{A. K.\} and Bugge, \{A. J.\}",

note = "Publisher Copyright: {\textcopyright} 2025 86th EAGE Annual Conference and Exhibition. All rights reserved.; 86th EAGE Annual Conference and Exhibition ; Conference date: 02-06-2025 Through 05-06-2025",

year = "2025",

doi = "10.3997/2214-4609.202510766",

language = "English",

series = "86th EAGE Annual Conference and Exhibition",

publisher = "European Association of Geoscientists and Engineers, EAGE",

booktitle = "86th EAGE Annual Conference and Exhibition",

}

Cui, Y, Huff, OR, Waheed, UB, Lie, JE, Evensen, AK & Bugge, AJ 2025, Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO. in 86th EAGE Annual Conference and Exhibition. 86th EAGE Annual Conference and Exhibition, European Association of Geoscientists and Engineers, EAGE, 86th EAGE Annual Conference and Exhibition, Toulouse, France, 2/06/25. https://doi.org/10.3997/2214-4609.202510766

Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO. / Cui, Y.; Huff, O. R.; Waheed, U. B. et al.
86th EAGE Annual Conference and Exhibition. European Association of Geoscientists and Engineers, EAGE, 2025. (86th EAGE Annual Conference and Exhibition).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO

AU - Cui, Y.

AU - Huff, O. R.

AU - Waheed, U. B.

AU - Lie, J. E.

AU - Evensen, A. K.

AU - Bugge, A. J.

PY - 2025

Y1 - 2025

N2 - This study aims to develop and validate a novel neural network architecture, KFNO, for reconstructing missing near-offset traces in marine seismic data. By integrating the Fourier Neural Operator (FNO) and the Kolmogorov-Arnold Network (KAN), the KFNO model aims to address the positional disparities between source and receiver lines that lead to missing traces, enhancing the accuracy and reliability of seismic data processing and interpretation.

AB - This study aims to develop and validate a novel neural network architecture, KFNO, for reconstructing missing near-offset traces in marine seismic data. By integrating the Fourier Neural Operator (FNO) and the Kolmogorov-Arnold Network (KAN), the KFNO model aims to address the positional disparities between source and receiver lines that lead to missing traces, enhancing the accuracy and reliability of seismic data processing and interpretation.

UR - https://www.scopus.com/pages/publications/105035324757

U2 - 10.3997/2214-4609.202510766

DO - 10.3997/2214-4609.202510766

M3 - Conference contribution

AN - SCOPUS:105035324757

T3 - 86th EAGE Annual Conference and Exhibition

BT - 86th EAGE Annual Conference and Exhibition

PB - European Association of Geoscientists and Engineers, EAGE

T2 - 86th EAGE Annual Conference and Exhibition

Y2 - 2 June 2025 through 5 June 2025

ER -

Marine Seismic Near Offset Data Reconstruction Using a KAN-Powered FNO

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this