US 12,313,800 B2
Ground roll attenuation using unsupervised deep learning
Haibin Di, Houston, TX (US); Nicolae Moldoveanu, Houston, TX (US); Hiren Maniar, Houston, TX (US); and Aria Abubakar, Sugar Land, TX (US)
Assigned to SCHLUMBERGER TECHNOLOGY CORPORATION, Sugar Land, TX (US)
Appl. No. 17/907,003
Filed by Schlumberger Technology Corporation, Sugar Land, TX (US)
PCT Filed Mar. 22, 2021, PCT No. PCT/US2021/023415
§ 371(c)(1), (2) Date Sep. 22, 2022,
PCT Pub. No. WO2021/194933, PCT Pub. Date Sep. 30, 2021.
Claims priority of provisional application 62/993,817, filed on Mar. 24, 2020.
Prior Publication US 2023/0109902 A1, Apr. 13, 2023
Int. Cl. G01V 1/36 (2006.01); G01V 1/34 (2006.01); G06N 3/0455 (2023.01); G06N 3/048 (2023.01); G06T 5/70 (2024.01)
CPC G01V 1/36 (2013.01) [G01V 1/345 (2013.01); G06N 3/0455 (2023.01); G06N 3/048 (2023.01); G06T 5/70 (2024.01); G01V 2210/32 (2013.01); G01V 2210/74 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20212 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for attenuating noise in seismic data, comprising:
receiving a seismic image;
generating a first image using a first neural network configured to identify low frequency noise in the seismic image;
generating a second image using a second neural network configured to identify reflections in the seismic image and to attenuate the low frequency noise;
generating a combined image by combining the first image and the second image; and
adjusting the first neural network and the second neural network to reduce a difference between the combined image and the seismic image using frequency constraint to guide separation of the seismic image into the first image and the second image.