Skip to Main Content
Skip Nav Destination

Deep learning for denoising

Yu Siwei, Ma Jianwei and Wang Wenlong
Deep learning for denoising
Geophysics (November 2019) 84 (6): V333-V350


Compared with traditional seismic noise attenuation algorithms that depend on signal models and their corresponding prior assumptions, removing noise with a deep neural network is trained based on a large training set in which the inputs are the raw data sets and the corresponding outputs are the desired clean data. After the completion of training, the deep-learning (DL) method achieves adaptive denoising with no requirements of (1) accurate modelings of the signal and noise or (2) optimal parameters tuning. We call this intelligent denoising. We have used a convolutional neural network (CNN) as the basic tool for DL. In random and linear noise attenuation, the training set is generated with artificially added noise. In the multiple attenuation step, the training set is generated with the acoustic wave equation. The stochastic gradient descent is used to solve the optimal parameters for the CNN. The runtime of DL on a graphics processing unit for denoising has the same order as the f-x deconvolution method. Synthetic and field results indicate the potential applications of DL in automatic attenuation of random noise (with unknown variance), linear noise, and multiples.

ISSN: 0016-8033
EISSN: 1942-2156
Coden: GPYSA7
Serial Title: Geophysics
Serial Volume: 84
Serial Issue: 6
Title: Deep learning for denoising
Affiliation: Harbin Institute of Technology, School of Mathematics, Harbin, China
Pages: V333-V350
Published: 201911
Text Language: English
Publisher: Society of Exploration Geophysicists, Tulsa, OK, United States
References: 63
Accession Number: 2020-011161
Categories: Applied geophysics
Document Type: Serial
Bibliographic Level: Analytic
Illustration Description: illus. incl. 1 table
Country of Publication: United States
Secondary Affiliation: GeoRef, Copyright 2020, American Geosciences Institute. Reference includes data from GeoScienceWorld, Alexandria, VA, United States. Reference includes data supplied by Society of Exploration Geophysicists, Tulsa, OK, United States
Update Code: 202008
Close Modal

or Create an Account

Close Modal
Close Modal