Seismic horizon picking using an artificial neural network

In seismic data interpretation, horizon picking is important for structural analysis, feature recognition, and site appraisal. However, horizon picking is still commonly done by hand, a process which is error prone and time consuming. Attempts to automate horizon picking are hindered by the absence of a clear, robust, and universal picking algorithm. A new method which combines a traditional approach to horizon picking with a new technique using a trained artificial neural network is presented. It is shown that this method makes better use of the general properties of horizons, is more robust than conventional pattern recognition techniques, and facilitates a solution to the problem of tracking through conventionally difficult regions containing faulting and other geophysical anomalies, where horizons are discontinuous