Method Of Weak Seismic Signal Recovery Based On The 3D-curvelet Transform. Computer Engineering And Applications

With the development of exploration technology in oil and gas,the exploitation technology of structural oil-gas reservoir has becoming matured.According to the analysis of seismic data,lithologic oil-gas reservoir was found under the stratum.This kind of oil-gas reservoir are small and deep,so it is not easy to identify oil and gas in the seismic data,and is easily disturbed by the noise.In order to localization these reservoirs more accurately,it is necessary to deal with the seismic data more accurately,such as reducing the noise interference,raising the formation information and so on.However,there are various defects in the traditional methods: The method based on Fourier transforms calculated average in the entire time domain and it will make the information of time loosen.The multi-scale and time-frequency analysis methods have good localization property in both time domain and frequency domain.But it cannot express the edge information of multidimensional signal well.In order to solve the above problems,a method based on parallel 3D-curvelet transform to recovery the weak signal of 3D seismic data is proposed in this paper.1.Decomposed the 3D seismic signal into sub-signal by the three-dimensional curvelet transform;2.Get adaptive threshold for denoising by the disadvantages of noise threshold and GCV;3.Improve the energy of signal by handling denoised signal in adaptive spectral whitening;4.Reconstruct the recovered data.So that we can recovery weak signal.In order to verify the effectiveness of our method,the contrast experiments were designed.And the denoising effect is evaluated by measuring the objective parameters.1.Compared with the method of wavelet transform,the experimental results show that the geological horizon became more clearly by using our method.2.Compared with the results of curvelet transform,it shows that our method can obtain more accurate results.The method proposed in this paper has enormous potential to improve the resolution of weak seismic signal.