Application of feedback connection artificial neural network to seismic data filtering

TL;DR

This paper explores using feedback neural networks, specifically Elman networks, for seismic data filtering, demonstrating their effectiveness in removing noise from synthetic and real seismic signals with simple training procedures.

Contribution

It introduces an Elman neural network approach for seismic data filtering, highlighting its simplicity and effectiveness compared to traditional methods.

Findings

Effective noise reduction in synthetic seismic data

Successful application to real seismic data

Network performance validated with cross-validation

Abstract

The Elman artificial neural network (ANN) (feedback connection) was used for seismic data filtering. The recurrent connection that characterizes this network offers the advantage of storing values from the previous time step to be used in the current time step. The proposed structure has the advantage of training simplicity by a back-propagation algorithm (steepest descent). Several trials were addressed on synthetic (with 10% and 50% of random and Gaussian noise) and real seismic data using respectively 10 to 30 neurons and a minimum of 60 neurons in the hidden layer. Both an iteration number up to 4000 and arrest criteria were used to obtain satisfactory performances. Application of such networks on real data shows that the filtered seismic section was efficient. Adequate cross-validation test is done to ensure the performance of network on new data sets.