基于贝叶斯反演的叠前数据同时插值和保幅去噪方法研究

Research on simultaneous interpolation and amplitude-preserving denoising based on Bayesian inversion of prestack data

近年来,地震勘探逐渐面向深层、复杂构造和岩性油气藏,与此同时,对地震资料处理也提出了更高的标准和要求.保幅降噪是地震资料处理中的重要环节.叠前地震资料的品质直接影响后续反演结果的精度,但道编辑后的实际数据除信噪比较低之外还存在缺道等现象,常规的保幅去噪方法无法同时进行道插值.本文基于贝叶斯反演框架提出了一种能够同时进行道插值的叠前地震资料保幅去噪方法.该方法充分考虑了现有方法的局限性,利用插值算子对缺失地震道进行插值处理,同时基于地震振幅随角度变化的连续性在差分域采用贝叶斯框架引入了先验约束,能够在保幅去噪的同时,进行道插值,极大改善了叠前地震资料的品质.合成和野外的例子均验证了该方法的有效性.

The target of seismic exploration gradually develops to deep,complex structure and lithology reservoir,which puts forward higher standards and requirements for seismic data processing.Amplitude-preserving denoising is an important step in seismic data processing.The quality of prestack seismic data directly affects the accuracy of the inversion results.However,in addition to the low signal-to-noise ratio,the field data after trace editing is still missing.The conventional amplitude-preserving denoising method cannot simultaneously perform trace interpolation.In this paper,we propose a prestack seismic data simultaneous interpolation and amplitude-preserving denoising method based on Bayesian inversion.Considering the limitations of the existing methods,the proposed method introduces the interpolation operator to interpolate the missing seismic traces.At the same time,due to the continuity of the seismic amplitude variation with angle,a priori constraint is introduced in the differential domain using the Bayesian framework.The method we proposed can achieve trace interpolation while amplitude-preserving denoising,which greatly improves the quality of prestack seismic data.Both synthetic and field examples verify the effectiveness of the method.