FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312111151> ?p ?o ?g. }

• W4312111151 abstract "Noise is ever present in seismic data and arises from numerous sources and is continually evolving, both spatially and temporally. The use of supervised deep learning procedures for denoising of seismic datasets often results in poor performance: this is due to the lack of noise-free field data to act as training targets and the large difference in characteristics between synthetic and field datasets. Self-supervised, blind-spot networks typically overcome these limitation by training directly on the raw, noisy data. However, such networks often rely on a random noise assumption, and their denoising capabilities quickly decrease in the presence of even minimally-correlated noise. Extending from blind-spots to blind-masks has been shown to efficiently suppress coherent noise along a specific direction, but it cannot adapt to the ever-changing properties of noise. To preempt the network’s ability to predict the signal and reduce its opportunity to learn the noise properties, we propose an initial, supervised training of the network on a frugally-generated synthetic dataset prior to fine-tuning in a self-supervised manner on the field dataset of interest. Considering the change in peak signal-to-noise ratio, as well as the volume of noise reduced and signal leakage observed, using a semi-synthetic example we illustrate the clear benefit in initialising the self-supervised network with the weights from a supervised base-training. This is further supported by a test on a field dataset where the fine-tuned network strikes the best balance between signal preservation and noise reduction. Finally, the use of the unrealistic, frugally-generated synthetic dataset for the supervised base-training includes a number of benefits: minimal prior geological knowledge is required, substantially reduced computational cost for the dataset generation, and a reduced requirement of re-training the network should recording conditions change, to name a few. Such benefits result in a robust denoising procedure suited for long term, passive seismic monitoring." @default.
• W4312111151 created "2023-01-04" @default.
• W4312111151 creator A5032021877 @default.
• W4312111151 creator A5071766835 @default.
• W4312111151 date "2022-12-12" @default.
• W4312111151 modified "2023-10-14" @default.
• W4312111151 title "Transfer learning for self-supervised, blind-spot seismic denoising" @default.
• W4312111151 cites W1984323018 @default.
• W4312111151 cites W2022567153 @default.
• W4312111151 cites W2030207039 @default.
• W4312111151 cites W2054113582 @default.
• W4312111151 cites W2335515558 @default.
• W4312111151 cites W2396378976 @default.
• W4312111151 cites W2475005008 @default.
• W4312111151 cites W2496746433 @default.
• W4312111151 cites W2550078111 @default.
• W4312111151 cites W2895546528 @default.
• W4312111151 cites W2999519822 @default.
• W4312111151 cites W3033557345 @default.
• W4312111151 cites W3033655015 @default.
• W4312111151 cites W3082024589 @default.
• W4312111151 cites W3134394808 @default.
• W4312111151 cites W3156923668 @default.
• W4312111151 cites W3207967945 @default.
• W4312111151 cites W3212113934 @default.
• W4312111151 cites W4251658380 @default.
• W4312111151 cites W4255421341 @default.
• W4312111151 cites W4285411663 @default.
• W4312111151 cites W4287887655 @default.
• W4312111151 doi "https://doi.org/10.3389/feart.2022.1053279" @default.
• W4312111151 hasPublicationYear "2022" @default.
• W4312111151 type Work @default.
• W4312111151 citedByCount "3" @default.
• W4312111151 countsByYear W43121111512022 @default.
• W4312111151 countsByYear W43121111512023 @default.
• W4312111151 crossrefType "journal-article" @default.
• W4312111151 hasAuthorship W4312111151A5032021877 @default.
• W4312111151 hasAuthorship W4312111151A5071766835 @default.
• W4312111151 hasBestOaLocation W43121111511 @default.
• W4312111151 hasConcept C115961682 @default.
• W4312111151 hasConcept C119857082 @default.
• W4312111151 hasConcept C136389625 @default.
• W4312111151 hasConcept C153180895 @default.
• W4312111151 hasConcept C154945302 @default.
• W4312111151 hasConcept C160920958 @default.
• W4312111151 hasConcept C163294075 @default.
• W4312111151 hasConcept C199360897 @default.
• W4312111151 hasConcept C202444582 @default.
• W4312111151 hasConcept C2779843651 @default.
• W4312111151 hasConcept C33923547 @default.
• W4312111151 hasConcept C41008148 @default.
• W4312111151 hasConcept C50644808 @default.
• W4312111151 hasConcept C9652623 @default.
• W4312111151 hasConcept C99498987 @default.
• W4312111151 hasConceptScore W4312111151C115961682 @default.
• W4312111151 hasConceptScore W4312111151C119857082 @default.
• W4312111151 hasConceptScore W4312111151C136389625 @default.
• W4312111151 hasConceptScore W4312111151C153180895 @default.
• W4312111151 hasConceptScore W4312111151C154945302 @default.
• W4312111151 hasConceptScore W4312111151C160920958 @default.
• W4312111151 hasConceptScore W4312111151C163294075 @default.
• W4312111151 hasConceptScore W4312111151C199360897 @default.
• W4312111151 hasConceptScore W4312111151C202444582 @default.
• W4312111151 hasConceptScore W4312111151C2779843651 @default.
• W4312111151 hasConceptScore W4312111151C33923547 @default.
• W4312111151 hasConceptScore W4312111151C41008148 @default.
• W4312111151 hasConceptScore W4312111151C50644808 @default.
• W4312111151 hasConceptScore W4312111151C9652623 @default.
• W4312111151 hasConceptScore W4312111151C99498987 @default.
• W4312111151 hasLocation W43121111511 @default.
• W4312111151 hasLocation W43121111512 @default.
• W4312111151 hasLocation W43121111513 @default.
• W4312111151 hasLocation W43121111514 @default.
• W4312111151 hasOpenAccess W4312111151 @default.
• W4312111151 hasPrimaryLocation W43121111511 @default.
• W4312111151 hasRelatedWork W2971361125 @default.
• W4312111151 hasRelatedWork W3046775127 @default.
• W4312111151 hasRelatedWork W3091943846 @default.
• W4312111151 hasRelatedWork W3094076422 @default.
• W4312111151 hasRelatedWork W3162567751 @default.
• W4312111151 hasRelatedWork W4220686584 @default.
• W4312111151 hasRelatedWork W4221088574 @default.
• W4312111151 hasRelatedWork W4285260836 @default.
• W4312111151 hasRelatedWork W4319309271 @default.
• W4312111151 hasRelatedWork W2092619848 @default.
• W4312111151 hasVolume "10" @default.
• W4312111151 isParatext "false" @default.
• W4312111151 isRetracted "false" @default.
• W4312111151 workType "article" @default.