Research on Seismic Noise Attenuation Method Based on Wavelet Transform and f-x-VMD

Random noise in seismic data seriously affects velocity analysis and migration imaging quality. Effective attenuation of random noise is an important link in seismic data processing. The traditional random noise attenuation method needs to select appropriate processing parameters, which has poor adaptability and requires a lot of calculation, resulting in the random noise can not be effectively attenuated. This paper combines f-x Variational Mode Decomposition (f-x-VMD) with adaptive wavelet threshold noise attenuation method and proposes a seismic noise attenuation method based on wavelet transform and f-x-VMD. In order to better attenuate the random noise, the signal is first converted to the frequency domain and is decomposed into intrinsic mode functions (IMFs) by VMD, then the IMFs are attenuated by the wavelet threshold noise attenuation method in the frequency domain and the time domain respectively, and finally the IMFs are reconstructed as signal through the inverse wavelet transform. The experimental results of simulated seismic data and real seismic data show that the proposed method has better random noise attenuation effect than the traditional method.