Resolution preserving structural steering for seismic data

Abstract

In this article, we will describe a new seismic filter to improve seismic data quality. This filter refers to the structural information and works to enhance the continuity of seismic horizons. Conventionally, the structural information is extracted through picking horizons and mapping them by interpreter. Recently alternative ways to extract the structural information directly from seismic data without manual horizon picking was developed. We applied this technique to estimate the spatially-variable local slope of the seismic events as the structural attribute by using the planewave destructor method. Attenuation of random noises and enhancement of structural continuity result in improving the quality of seismic interpretation. However random noise attenuation and preserving spatial resolution always have the trade-off relation in seismic data processing. To overcome this problem, we applied the “symmetric nearest neighbor” method. The symmetric nearest neighbor filtering is widely applied in the field of image processing as an edge-preserving smoothing filter for square dataset. We applied this method to the segmental dataset in seismic data processing.

Numerical tests on the field seismic data showed that this filter efficiently attenuated random noises while preserving the spatial resolution of the seismic image.