Noise Attenuation Technique Using the 2D Wavelet Transform and the Adaptive Deconvolution on Pre-Stack and Post-Stack data
Lucas José A. de Almeida, Milton José Porsani · 2013
The wavelet transform uses wavelets in order to perform a spectral decomposition on seismic data, segregating it in several sub-bands. These sub-bands are oriented and contain somewhat precise frequency information, which depends on the wavelet that was used to apply the transform (Cohen, 1993). Since the ground roll is a type of seismic noise that is mostly vertically oriented, it will be separated from the rest of the non-vertical seismic events, making it easier to filter it without altering other seismic events of interest, such as the reflections. In this paper, the wavelet transform is used as a mean to segregate the ground roll from the reflections, while an adaptive deconvolution is applied only on the band where the ground roll is represented, in order to filter it. We applied the method using two different approaches: pre-stack and post-stack data. The results show that the post-stack application provided a better image.