Frequency domain self-adaptive non-linear earthquake imaging filtering method

Abstract

A frequency domain self-adaptive non-linear earthquake imaging filtering method comprises the steps as follows: firstly, earthquake data are input for calculation and analysis of frequency amplitude spectrum, original earthquake data are subjected to wavelet transformation scale component decomposition with a wavelet basis, components of the earthquake data on different scales are obtained, frequency domain transformation of the earthquake data is performed according to the corresponding relation between the scales and the frequency, single-frequency component earthquake data are extracted from the frequency domain earthquake data subjected to wavelet transformation, the single-frequency component earthquake data in which noise are mainly distributed are subjected to edge function solution of an improved self-adaptive non-linear diffusion equation, finally, all frequency component data can meet the requirement of the maximum signal-to-noise ratio, signal-to-noise ratio quality monitoring is performed on the iterative process of repeated self-adaptive non-linear diffusion filtering, so that the optimal iteration times of filtering is determined, all frequency component earthquake data after repeated iteration filtering are subjected to wavelet reconstruction synthesis, and the final earthquake data after imaging filtering are obtained.

Description

technical field [0001] The invention belongs to the technical field of seismic data processing, and in particular relates to a frequency domain self-adaptive nonlinear seismic imaging filtering method. Background technique [0002] In the process of seismic imaging, faults, cracks, pinch-outs, geological anomalies, etc. will cause changes in the details of seismic image edges, textures, etc. At the same time, due to imaging reasons in the process of seismic wave acquisition and processing, these anomalies will inevitably carry strong noise . The difficulty in dealing with such anomalies lies in maintaining the authenticity of seismic image details and effectively removing noise. [0003] At present, wavelet transform or anisotropic diffusion filtering are mainly used alone. The wavelet threshold method is fast in calculation but difficult to select the threshold; the image denoising effect obtained by the anisotropic diffusion method is good, but it is easy to cause the ima...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that in the case of complex formations, when the noise is large, the processing effect on high-frequency noise signals is often not good for low-frequency component noise.

The disadvantage of this method is that the seismic image processing in the scale domain does not consider the actual frequency domain information of the seismic data, and the PM diffusion equation is directly used. The small texture changes caused by the detailed geological conditions are likely to cause the boundary of the geological structure obtained after processing to differ from A sick image of real error

Images