FIELD: seismic data processing.
SUBSTANCE: invention relates to methods for processing and interpreting seismic data to determine tectonic faults. Seismic 3D images of the earth's crust are obtained. The resulting images are transmitted to the input of the trained neural network. At the output of the trained neural network, a binary mask of 3D seismic images is obtained with the prediction of tectonic faults. Post-processing of the obtained binary mask of 3D seismic images of tectonic faults is performed.
EFFECT: automation of the process of determining tectonic faults.
1 cl, 8 dwg
