English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Noise suppression is an essential step in any seismic processing workflow. Aportion of this noise, particularly in land datasets, presents itself as randomnoise. In recent years, neural networks have been successfully used to denoiseseismic data in a supervised fashion. However, supervised learning always comeswith the often unachievable requirement of having noisy-clean data pairs fortraining. Using blind-spot networks, we redefine the denoising task as aself-supervised procedure where the network uses the surrounding noisy samplesto estimate the noise-free value of a central sample. Based on the assumptionthat noise is statistically independent between samples, the network strugglesto predict the noise component of the sample due to its randomnicity, whilstthe signal component is accurately predicted due to its spatio-temporalcoherency. Illustrated on synthetic examples, the blind-spot network is shownto be an efficient denoiser of seismic data contaminated by random noise withminimal damage to the signal; therefore, providing improvements in both theimage domain and down-the-line tasks, such as inversion. To conclude the study,the suggested approach is applied to field data and the results are comparedwith two commonly used random denoising techniques: FX-deconvolution andCurvelet transform. By demonstrating that blind-spot networks are an efficientsuppressor of random noise, we believe this is just the beginning of utilisingself-supervised learning in seismic applications.

The potential of self-supervised networks for random noise suppression in seismic data