The Denosing of Seismic Signal Based on Generalized S Transform

The S transform filtering methods are restricted by a fixed window function, so generalized S transform filtering methods are put forward. Firstly, generalized S transform whose time frequency resolution can be adjustable is achieved by altering normal S transform. The numerical implementation of generalized S transform is introduced later. Secondly, generalized S transform filtering methods are applied to denoising of seismic signal, respectively use time-frequency filtering and threshold filtering based on generalized S transform to filter, the filtering operators are designed for the different frequency characteristics of signals. By synthetic seismic signals and real seismic data simulation demonstrate that generalized S transform filtering methods have obvious effectiveness, the methods eliminate effectively noises at different time intervals and different frequencies, reserve the primary information maximally. This paper shows that generalized S transform filtering methods are effective and provide a effective method for the denoising of seismic signal.