DEEP NEURAL NETWORK-BASED AIRBORNE EM DATA INVERSION SUITABLE FOR MOUNTAINOUS FIELD SITES
- Authors
- Bang, Minkyu; Byun, Joongmoo; Seol, Soon Jee
- Issue Date
- Jun-2022
- Publisher
- European Association of Geoscientists and Engineers, EAGE
- Citation
- 83rd EAGE Conference and Exhibition 2022, v.2, pp.1288 - 1292
- Indexed
- SCOPUS
- Journal Title
- 83rd EAGE Conference and Exhibition 2022
- Volume
- 2
- Start Page
- 1288
- End Page
- 1292
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/173244
- DOI
- 10.3997/2214-4609.202210341
- ISSN
- 0000-0000
- Abstract
- Deep neural network (DNN) based inversion of airborne electromagnetic (AEM) data can show similar performance to the conventional inversion and can greatly reduce the computation time. However, interpretation techniques using a neural network are still early stage of field application, so there is no study which can consider the topographical changes. Since AEM surveys are usually conducted in mountainous area, topographical variations can cause the distortion in EM data. Thus, in this study, we proposed an idea that can consider the topography information when interpreting AEM data using a DNN. We confirmed the performance of the trained DNN through a numerical experiment and verified its validity by applying it to the field dataset.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 자원환경공학과 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.