Research Paper

Split Viewer

Econ. Environ. Geol. 2024; 57(1): 51-71

Published online February 29, 2024

https://doi.org/10.9719/EEG.2024.57.1.51

© THE KOREAN SOCIETY OF ECONOMIC AND ENVIRONMENTAL GEOLOGY

Study on the Seismic Random Noise Attenuation for the Seismic Attribute Analysis

Jongpil Won1, Jungkyun Shin2, Jiho Ha2, Hyunggu Jun1,*

1Department of Geology, Kyungpook National University, Daegu 41566, Republic of Korea
2Pohang Branch, Korea Institute of Geoscience and Mineral Resources (KIGAM), Pohang 37559, Republic of Korea

Correspondence to : *hgjun@knu.ac.kr

Received: September 15, 2023; Revised: October 24, 2023; Accepted: October 24, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided original work is properly cited.

Abstract

Seismic exploration is one of the widely used geophysical exploration methods with various applications such as resource development, geotechnical investigation, and subsurface monitoring. It is essential for interpreting the geological characteristics of subsurface by providing accurate images of stratum structures. Typically, geological features are interpreted by visually analyzing seismic sections. However, recently, quantitative analysis of seismic data has been extensively researched to accurately extract and interpret target geological features. Seismic attribute analysis can provide quantitative information for geological interpretation based on seismic data. Therefore, it is widely used in various fields, including the analysis of oil and gas reservoirs, investigation of fault and fracture, and assessment of shallow gas distributions. However, seismic attribute analysis is sensitive to noise within the seismic data, thus additional noise attenuation is required to enhance the accuracy of the seismic attribute analysis. In this study, four kinds of seismic noise attenuation methods are applied and compared to mitigate random noise of poststack seismic data and enhance the attribute analysis results. FX deconvolution, DSMF, Noise2Noise, and DnCNN are applied to the Youngil Bay high-resolution seismic data to remove seismic random noise. Energy, sweetness, and similarity attributes are calculated from noise-removed seismic data. Subsequently, the characteristics of each noise attenuation method, noise removal results, and seismic attribute analysis results are qualitatively and quantitatively analyzed. Based on the advantages and disadvantages of each noise attenuation method and the characteristics of each seismic attribute analysis, we propose a suitable noise attenuation method to improve the result of seismic attribute analysis.

Keywords seismic attribute analysis, noise attenuation, machine learning, gas distribution, seismic data

탄성파 속성 분석을 위한 탄성파 자료 무작위 잡음 제거 연구

원종필1 · 신정균2 · 하지호2 · 전형구1,*

1경북대학교 지질학과
2한국지질자원연구원 포항지질자원실증연구센터

요 약

탄성파 탐사는 지하자원 개발, 지반 조사, 지층 모니터링 등에 널리 사용되고 있는 지구물리탐사 방법으로 정확한 지층 구조 영상을 제공해주기 때문에 지층의 지질학적 특성 해석에 필수적으로 활용된다. 일반적으로는 탄성파 구조 보정 영상을 시각적으로 분석하여 지질학적 특성을 해석하지만 최근에는 탄성파 구조 보정 자료에 대한 정량적인 분석을 통해 원하는 지질학적 특성을 정확하게 추출하고 해석하는 탄성파 속성 분석이 널리 연구되고 있다. 탄성파 속성 분석은 탄성파 자료에 기반한 지질학적 해석에 정량적인 근거를 제시해줄 수 있기 때문에 석유 및 가스 저류층 분석, 단층 및 균열대 조사, 지층 가스 분포 파악 등의 다양한 분야에서 활용되고 있다. 하지만 탄성파 속성 분석은 탄성파 자료 내 잡음에 취약하므로 속성 분석의 정확도 향상을 위해서는 중합 후 탄성파 자료에 대한 추가적인 잡음 제거가 수반되어야 한다. 본 연구에서는 중합 후 탄성파 자료에 대한 무작위 잡음 제거 및 및 탄성파 속성 분석 정확도 개선을 위해 4가지의 잡음 제거 방법을 적용하고 비교한다. FX 디콘볼루션,DSMF, Noise2Noiose, DnCNN을 각각 포항 영일만 고해상 탄성파 자료에 적용하여 탄성파 무작위 잡음을 제거하고 잡음이 제거된 탄성파 자료로부터 에너지, 스위트니스, 유사도 속성을 계산한다. 그리고 각 잡음 제거 방법의 특성, 잡음 제거 결과, 탄성파 속성 분석 결과를 정성적 및 정량적으로 분석한 후, 이를 기반으로 탄성파 속성 분석 결과 향상을 위한 최적의 잡음 제거 방법을 제안한다.

주요어 탄성파 속성 분석, 잡음 제거, 기계 학습, 가스 분포, 탄성파 자료

Article

Research Paper

Econ. Environ. Geol. 2024; 57(1): 51-71

Published online February 29, 2024 https://doi.org/10.9719/EEG.2024.57.1.51

Copyright © THE KOREAN SOCIETY OF ECONOMIC AND ENVIRONMENTAL GEOLOGY.

Study on the Seismic Random Noise Attenuation for the Seismic Attribute Analysis

Jongpil Won1, Jungkyun Shin2, Jiho Ha2, Hyunggu Jun1,*

1Department of Geology, Kyungpook National University, Daegu 41566, Republic of Korea
2Pohang Branch, Korea Institute of Geoscience and Mineral Resources (KIGAM), Pohang 37559, Republic of Korea

Correspondence to:*hgjun@knu.ac.kr

Received: September 15, 2023; Revised: October 24, 2023; Accepted: October 24, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided original work is properly cited.

Abstract

Seismic exploration is one of the widely used geophysical exploration methods with various applications such as resource development, geotechnical investigation, and subsurface monitoring. It is essential for interpreting the geological characteristics of subsurface by providing accurate images of stratum structures. Typically, geological features are interpreted by visually analyzing seismic sections. However, recently, quantitative analysis of seismic data has been extensively researched to accurately extract and interpret target geological features. Seismic attribute analysis can provide quantitative information for geological interpretation based on seismic data. Therefore, it is widely used in various fields, including the analysis of oil and gas reservoirs, investigation of fault and fracture, and assessment of shallow gas distributions. However, seismic attribute analysis is sensitive to noise within the seismic data, thus additional noise attenuation is required to enhance the accuracy of the seismic attribute analysis. In this study, four kinds of seismic noise attenuation methods are applied and compared to mitigate random noise of poststack seismic data and enhance the attribute analysis results. FX deconvolution, DSMF, Noise2Noise, and DnCNN are applied to the Youngil Bay high-resolution seismic data to remove seismic random noise. Energy, sweetness, and similarity attributes are calculated from noise-removed seismic data. Subsequently, the characteristics of each noise attenuation method, noise removal results, and seismic attribute analysis results are qualitatively and quantitatively analyzed. Based on the advantages and disadvantages of each noise attenuation method and the characteristics of each seismic attribute analysis, we propose a suitable noise attenuation method to improve the result of seismic attribute analysis.

Keywords seismic attribute analysis, noise attenuation, machine learning, gas distribution, seismic data

탄성파 속성 분석을 위한 탄성파 자료 무작위 잡음 제거 연구

원종필1 · 신정균2 · 하지호2 · 전형구1,*

1경북대학교 지질학과
2한국지질자원연구원 포항지질자원실증연구센터

Received: September 15, 2023; Revised: October 24, 2023; Accepted: October 24, 2023

요 약

탄성파 탐사는 지하자원 개발, 지반 조사, 지층 모니터링 등에 널리 사용되고 있는 지구물리탐사 방법으로 정확한 지층 구조 영상을 제공해주기 때문에 지층의 지질학적 특성 해석에 필수적으로 활용된다. 일반적으로는 탄성파 구조 보정 영상을 시각적으로 분석하여 지질학적 특성을 해석하지만 최근에는 탄성파 구조 보정 자료에 대한 정량적인 분석을 통해 원하는 지질학적 특성을 정확하게 추출하고 해석하는 탄성파 속성 분석이 널리 연구되고 있다. 탄성파 속성 분석은 탄성파 자료에 기반한 지질학적 해석에 정량적인 근거를 제시해줄 수 있기 때문에 석유 및 가스 저류층 분석, 단층 및 균열대 조사, 지층 가스 분포 파악 등의 다양한 분야에서 활용되고 있다. 하지만 탄성파 속성 분석은 탄성파 자료 내 잡음에 취약하므로 속성 분석의 정확도 향상을 위해서는 중합 후 탄성파 자료에 대한 추가적인 잡음 제거가 수반되어야 한다. 본 연구에서는 중합 후 탄성파 자료에 대한 무작위 잡음 제거 및 및 탄성파 속성 분석 정확도 개선을 위해 4가지의 잡음 제거 방법을 적용하고 비교한다. FX 디콘볼루션,DSMF, Noise2Noiose, DnCNN을 각각 포항 영일만 고해상 탄성파 자료에 적용하여 탄성파 무작위 잡음을 제거하고 잡음이 제거된 탄성파 자료로부터 에너지, 스위트니스, 유사도 속성을 계산한다. 그리고 각 잡음 제거 방법의 특성, 잡음 제거 결과, 탄성파 속성 분석 결과를 정성적 및 정량적으로 분석한 후, 이를 기반으로 탄성파 속성 분석 결과 향상을 위한 최적의 잡음 제거 방법을 제안한다.

주요어 탄성파 속성 분석, 잡음 제거, 기계 학습, 가스 분포, 탄성파 자료

    Fig 1.

    Figure 1.Schematic diagrams of seismic random noise attenuation based on the (a) typical supervised learning and (b) Noise2Noise unsupervised learning.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 2.

    Figure 2.Network architecture of Noise2Noise used in this study.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 3.

    Figure 3.Network architecture of DnCNN used in this study.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 4.

    Figure 4.Poststack time migration cube of the Youngil Bay 3D high-resolution seismic data.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 5.

    Figure 5.The 89th inline section of poststack time migration data before denoising.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 6.

    Figure 6.Denoised result and estimated noise of FX deconvolution ((a) and (b)), DSMF ((c) and (d)), Noise2Noise ((e) and (f)) and DnCNN ((g) and (h)).
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 7.

    Figure 7.The energy attribute result (a) before noise attenuation and after applying (b) FX deconvolution, (c) DSMF, (d) Noise2Noise, (e) DnCNN.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 8.

    Figure 8.The sweetness attribute result (a) before noise attenuation and after applying (b) FX deconvolution, (c) DSMF, (d) Noise2Noise, (e) DnCNN.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 9.

    Figure 9.The similarity attribute result (a) before noise attenuation and after applying (b) FX deconvolution, (c) DSMF, (d) Noise2Noise, (e) DnCNN.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Fig 10.

    Figure 10.The frequency spectrum of the 89th inline poststack seismic section before and after noise attenuation.
    Economic and Environmental Geology 2024; 57: 51-71https://doi.org/10.9719/EEG.2024.57.1.51

    Table 1 . Comparison of the four types of denoising method.

    Application difficultyComputation timeSignal to Noise ratioSignal distortion
    FX deconvolutionLow1m 24s6.97Medium
    DSMFMedium2h 9m 18s11.28High
    Noise2NoiseHigh1h 21 33s7.66High
    DnCNNHigh2h 44m 1s10.25Low

    KSEEG
    Apr 30, 2024 Vol.57 No.2, pp. 107~280

    Stats or Metrics

    Share this article on

    • kakao talk
    • line

    Related articles in KSEEG

    Economic and Environmental Geology

    pISSN 1225-7281
    eISSN 2288-7962
    qr-code Download