Seismic wavelet extraction using artificial neural networks

Geophysical events are of interest to the interpreter as an indicator of geological boundaries and structures. In structural analysis, extraction of reflection events is still commonly done by hand, a process which is error-prone and time consuming. Attempts to automate the process are hindered by the absence of a clear, robust and universal picking algorithm. A new feature extraction technique for seismic data interpretation, using a trained Artificial Neural Network is presented. It is shown that this method is useful in extracting geophysical events where conventional pattern recognition techniques may fail.<
>