81st EAGE Conference and Exhibition 2019, Page. 1-5
Abstract
we propose an electrical resistivity imaging method from electromagnetic data based on the ever-evolving machine learning technique. This method is applied to delineate salt body that is essential for hydrocarbon reservoir imaging and uses a fully convolutional network to preserve the spatial information of the input data. The proposed network is trained using synthetic data and shows impressive results when applied to unseen test data.