Concurrent Detection of Salt Domes and Faults using ResNet with U-Net

Salt domes and faults are important seismic events and constitute potential indicators of hydrocarbon (i.e. oil, gas) accumulation. Seismic data interpretation is one fundamental process for identifying such events and even more. In the last decades a number of techniques have been proposed for automation of the seismic interpretation. The main goals is to speed up the interpretation process and improve interpretation accuracy. In this paper we present a brief overview of important approaches developed for salt and fault identification while categorizing them into feature engineering based and deep learning (DL) based. We present our DL framework for simultaneous salt domes and faults delineation with highlighting the promising preliminary results obtained trough applications to real world seismic datasets.