A systematic review of deep learning for structural geological interpretation

It is well known that seismic data (or seismic volumes/images) are one of the primary work materials of the oil and gas industry. Nevertheless, the manual interpretation of such data has become increasingly time-consuming and prone to errors. This problem arises due to the massive amount of data and the complexity and variability of geological patterns. Over the last few years, image processing methods from the Artificial Intelligence (AI) sub-field Deep Learning (DL) have become valuable tools for seismic volume interpretation. DL techniques are already incorporated in commercial seismic interpretation software such as GeoTeric and Earth Science Analytics. Overall, DL accelerates the seismic interpretation process and enhances the results’ precision and consistency. This allows for a better understanding of geological structures and reduces the risks associated with exploring and producing natural resources. Despite these facts, a systematic and broad review of DL techniques for different structural geological interpretation tasks has not been made available to researchers and practitioners. This article comprehensively reviews current techniques, models, and practices in DL-based seismic volume interpretation based on three different structural interpretation tasks: Fault interpretation, Horizon estimation, and Relative Geological Time (RGT) estimation, three complementary pillars of the geological framework. Our review is based on 85 relevant research articles, provides a foundational view of DL concepts, and introduces taxonomies for applying DL to structural geological interpretation. Our review also paves the way for further research by presenting the challenges encountered when applying deep learning in geological tasks and additional directions for future endeavors. The major applications of the techniques presented here are identifying architectural depositional elements and petrophysical properties of seismic images for natural oil and gas exploration.