Research On The Reconstruction Of Two-dimensional Seismic Signal Based On Morphological Component Analysis

Seismic signal reconstruction is of great significance in petroleum seismic data processing. Provide regular input data for the wave equation migration and other post-processing module. Achieve different block’s merging process. Helps to eliminate plume drifting and multiple waves at sea in the acquisition, etl. In order to better express different intrinsic characteristics in seismic signals. Further improve the performance of the reconstruction algorithm, Based on the morphological component analysis theoretical framework take a new research on the seismic signal reconstruction algorithm.The main research work in this paper is as follows:(1)Seismic signal reconstruction under the morphological component analysis framework combined with DCT and curvelet dictionary.Aiming at the problem that the mathematics transforms and dictionaries can not effectively depict the morphological features diversity of seismic signals, this paper proposes a seismic signal reconstruction method under the morphological component analysis(MCA) framework combined with discrete cosine transform(DCT) and curvelet double dictionary. Firstly the seismic signal reconstruction model is built under the MCA framework. And then the signal is decomposed into local singular and smooth linear component based on the model. Following that, the local singular component is represented by DCT dictionary, and the smooth linear component is represented by curvelet dictionary. We combine two kinds of components together after iterative reconstruction. The experiments on synthetic and real seismic signals illustrates that the proposed method can reconstruct signals very well. The reconstruct precision of the method is not only higher than some double dictionary combinations such as UDWT and curvelet, curvelet and curvelet, CMBF and curvelet, but also higher than some single dictionaries such as DCT, UDWT, or curvelet etc.(2) 2D Seismic Data Reconstruction based on Wave atom transform dictionarySeismic signals have the characteristics of turbulence, although DCT or UDWT transform along the direction of the shock morphological characteristics can be caught, but the direction character description is not enough, but wave atoms transform as a new multiscale geometric analysis tools, could make up for a lack of the above, and has the optimal sparse representation to the shock signal. So we come up with 2D seismic data reconstruction based on wave atom transform dictionary. The algorithm firstly to form 2D seismic signal component decomposition, obtain local singular component and smooth and linear component respectively, and then used wave atom dictionary to reconstruct the local singular component, curvelet dictionary is still used to reconstruct smooth and linear component. merge reconstruction results finally. Reconstruction experiments show that dictionary based on wave atoms transform of seismic signal reconstruction precision not only higher than that of DCT +curvelet dictionary, also is more than the UDWT+ curvelet dictionary、curvelet +curvelet dictionary and CMFB+curvelet dictionary or Other pairs of dictionary.(3) Seismic signal reconstruction under the morphological component analysis framework based on multi- dictionary linear combination.Seismic signal with multifarious features is the response to underground complex geological structure. The diversity of the its characteristics can not be portrayed effectively relying on one or two dictionaries. This paper attempts to achieve a finer decomposition of seismic signals under MCA framework and complete the reconstruction of seismic signals through a linear combination of multiple redundant dictionary. To reconstruct the components with different characteristics targetedly using hard threshold screening and decreasing index update strategy achieves a better performance in the reconstruction. Experiments show that the algorithm based on the linear combinations of multiple dictionaries adapts to the complex seismic signal reconstruction scenarios and reconstructs the seismic signal effectively.The algorithm performs fairly comparing with the algorithm based on wave atom transform dictionary. It is superior to the reconstruction effect using two dictionaries and more superior to that using one.