Smoothing Kinematic Wavefield Attributes to Reduce Random Noise and Enhance Signal-to-noise Ratio in Seismic Imaging

In seismic data, random noise can affect the continuity and identification of reflectors. In Common Reflection Surface (CRS) stacked data, part of this noise results from fluctuations and non-physical values of the estimated CRS parameters. This paper presents the result of a method to overcome this problem based on smoothing CRS parameters through the application of local statistics in small windows aligned with reflection events. Smoothing CRS parameters on a real marine data resulted in an increase of S/N ratio, with consequent highlight and greater continuity of the reflections.