Complete identification of complex salt geometries from inaccurate migrated subsurface offset gathers using deep learning

Abstract

Delimiting salt inclusions from migrated images is a time-consuming activity that relies on highly human-curated analysis and is subject to interpretation errors or limitations of the methods available. We develop the use of migrated images produced from an inaccurate velocity model (with a reasonable approximation of sediment velocity, but without salt inclusions) to predict the correct salt inclusions shape using a convolutional neural network. Our approach relies on subsurface common-image gathers to focus the sediments’ reflections around the zero offset and to spread the energy of salt reflections over large offsets. Using synthetic data, we train a U-Net to use common-offset subsurface images as input channels and the correct salt masks as the output of a semantic segmentation problem. The network learns to predict the salt inclusions masks with high accuracy; moreover, it also performs well when applied to synthetic benchmark data sets that were not previously introduced. Our training process successfully tunes the U-Net to learn the shape of complex salt bodies from partially focused subsurface offset images.