We have designed a deep-learning workflow to interactively track seismic geobodies. The algorithm is based on a flood-filling network, which performs iterative segmentation and moving the field of view (FoV). The proposed network takes the previous mask output, together with the seismic image in a new FoV, as a combined input to predict the mask at this FoV. The movement of the FoV is guided by the flood-filling algorithm to visit and segment the full extent of a geobody. Unlike conventional seismic image segmentation methods, the proposed workflow can not only detect geobodies, but it can also track individual geobody instances.
Paper
Full text
Interactively tracking seismic geobodies with a deep-learning flood-filling network
Semantic Scholar · Geology · 2021
Abstract
We have designed a deep-learning workflow to interactively track seismic geobodies. The algorithm is based on a flood-filling network, which performs iterative segmentation and moving the field of view (FoV). The proposed network takes the previous mask output, together with the seismic image in a new FoV, as a combined input to predict the mask at this FoV. The movement of the FoV is guided by the flood-filling algorithm to visit and segment the full extent of a geobody. Unlike conventional seismic image segmentation methods, the proposed workflow can not only detect geobodies, but it can also track individual geobody instances.