Hybrid, wavelet transform based, noise attenuation

Article type: Research Article

Authors: Zhang, Rongfeng^{; *} | Trad, Daniel^{; **} | Ulrych, Tadeusz J.

Affiliations: Department of Earth and Ocean Sciences, University of British Columbia, 2219 Main Mall, Vancouver, BC, Canada V6T 1Z4. E-mail:rzhang@eos.ubc.ca

Correspondence: [*] Corresponding author.

Note: [**] Present address: Veritas Geoservices, 2200-715 5thAvenue SW, Calgary, Alberta, Canada T2P 5A2; E-mail: daniel_trad@veritasdgc.com.

Abstract: Field seismic data often contain various types of strong or weak noise and interferences. If the noise is several orders of magnitude larger than the signal, most techniques applied in seismic data processing will be severely affected. Denoising methods, even very robust schemes such as physical wavelet frame denoising are no exception. In this paper, we present a robust, data adaptive and fast 1D wavelet transform method to attenuate this type of noise and develop a hybrid strategy for noise attenuation in combination with the 2D wavelet frame denoising.

DOI: 10.3233/ICA-2005-12107

Journal: Integrated Computer-Aided Engineering, vol. 12, no. 1, pp. 91-98, 2005