AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
![]() ![]() Missing well log prediction using a convolutional long short-term memory network Note: we compute and display all the results using our freely available Java packages (etc., Mines JTK). We then use the models to simulate numerous synthetic seismic datasets to train deep convolutional neural networks for recognizing real structural and stratigraphic features in field seismic datasets. We first perform forward numerical modeling to automatically generate numerous realistic structural and stratigraphic models. We currently focus on deep learning for seismic structural and stratigraphic interpretation. We further use the automatically extracted seismic structural and stratigraphic information, combined with well-log properties, to build subsurface models. For example, we develop algorithms to efficiently extract geologic features including faults, unconformities, horizons, channels, and salt boundaries from 3D seismic volumes. We work on computer-assisted interpretation of geophysical datasets including seismic images, well-logs, borehole images and so on. Research Computational interpretation group Image processing, Machine learning, Seismic interpretation, Subsurface modeling, Geophysical inversion, Computer vision, Computer graphics…. ![]()
0 Comments
Read More
Leave a Reply. |