Random Noise Reduction in Seismic Data by Using Bidimensional Empirical Mode Decomposition and Shearlet Transform

Abstract

Due to the limitation of the seismic data acquisition environment and instrument, seismic data are often subjected to random noise interference. At the same time, random noise is inevitably introduced in the processing of seismic data. To solve the problem, this paper proposes a seismic data denoising approach based on bidimensional empirical mode decomposition (BEMD) and shearlet transform. In the beginning, the BEMD is used to decompose the seismic data with noise, and several intrinsic modal functions (IMFs) are obtained according to the frequency distribution from high to low. Then, the shearlet transform is applied to reduce the noise of the high-frequency IMF. Finally, the denoised high-frequency IMF component and the low-frequency IMF component are accumulated and reconstructed to achieve the noise reduction of seismic data. The proposed method was used to perform noise reduction experiments on seismic data. The results show that the proposed method performs better and can reduce the uneven deformation of seismic wavefront compared with other methods.