Edge-preserving smoothing for seismic image using coherence-enhancing anisotropic diffusion

摘要

Currently, there are a variety of post-stack time-domain and frequency-domain de-noising method, but they will reduce the vertical and horizontal resolution of seismic data. Actually it is negative for analysis of geometric attribute and other follow-up work, so we need a filtering that can maintain the reflection angle as well as the contact relationship of strata. Coherence-enhancing diffusion filtering is just to meet the requirements, because it combines the smoothing and edge detection. The technique improves signal-to-noise ratio of seismic data as well as enhance the continuity of reflection events and protect important geological structure. In this paper, coherence-enhancing anisotropic diffusion filtering is used to process seismic image, and as a result it is really useful for interpretation of low S/N ratio seismic data. Examples show the remarkable effects of this operation.