Applicability of NDM-Net Simulated Data to Prestack Depth Migration | Sciact - Система учета научной деятельности ИМ СО РАН

Applicability of NDM-Net Simulated Data to Prestack Depth Migration Научная публикация

Конференция

The 26th International Conference on Computational Science and Its Applications
30 июн. - 3 июл. 2026 , Braga, University of Minho

Журнал

Lecture Notes in Computer Science
ISSN: 0302-9743 , E-ISSN: 1611-3349

Вых. Данные

Год: 2026, Том: 16761, Страницы: 58-69 Страниц : 12 DOI: 10.1007/978-3-032-30527-5_4

Авторы

Gondyul Elena ¹ , Dubrovina Valeria ² , Lisitsa Vadim ¹ , Vishnevsky Dmitry ¹

Организации

1	Sobolev Institute of Mathematics SB RAS, Novosibirsk, 630090, Russia
2	LLC PetroTrace, Moscow, 115114, Russia

Информация о финансировании (1)

Российский научный фонд

22-11-00004-П

We present an analysis of the applicability of previously reported approach (NDM-net) to the construction of depth seismic images. The NDM-net allows us to simulate seismic data by combining finite differences with deep learning method. Coarse computational grids are used for the fast generation of inaccurate datasets. After that, the training dataset is simulated for a small number of shot points on fine grids. Then, the NDM-net is trained to map inaccurate data to accurate common shot gathers. Previously, we illustrated the high quality of the generated shot gathers. In this research, we applied the prestack depth migration to estimate the effect of the NDM-net numerical errors on the resulting seismic images.

Gondyul E. , Dubrovina V. , Lisitsa V. , Vishnevsky D.
Applicability of NDM-Net Simulated Data to Prestack Depth Migration
Lecture Notes in Computer Science. 2026. V.16761. P.58-69. DOI: 10.1007/978-3-032-30527-5_4 Scopus OpenAlex

Опубликована online:

3 июл. 2026 г.

≡ Scopus:	2-s2.0-105044908558
≡ OpenAlex:	W7167092820