基于Curvelet变换的线杆共振干扰去除方法

Curvelet transform-based denoising of resonance interference induced by electrical poles in seismic exploration

线杆共振干扰是中浅层地震勘探常见干扰之一,尤其对浅部数据影响较大,由于石油、煤田勘探涉及此类干扰较少,缺乏相关研究内容。Curvelet变换可以获得图像平滑区域和边缘部分的稀疏表达,也能满足时变信号处理的要求,在地震资料处理中取得了较好的效果。本文根据线杆共振干扰在地震数据中表现的特点,提出了一种基于Curvelet变换的线杆共振干扰去除方法,首先通过分析线杆共振干扰与有效信息在Curvelet域的特征差异,借助Curvelet变换的多尺度、多方向特性实现波场分离,然后根据本文设计的非线性阈值函数对干扰系数进一步衰减。通过实际数据的应用分析,发现本文提出的方法可以有效地去除线杆共振干扰,同时可以较好地保护有效信号,去噪后资料的信噪比及分辨率均有不同程度的提高,从而证明了该方法的有效性。

The resonance interference induced by electrical poles is a common type of noise in middle-shallow seismic exploration,especially for shallow data.However,relevant studies are scarce since it is rarely involved in petroleum and coalfield exploration.The Curvelet transform allows for a sparse representation of smooth regions and edges in images and can meet the requirements of time-varying signal processing,thus achieving good effects in the processing of seismic data.Based on the characteristics of the electrical pole-induced resonance interference in seismic data,this paper proposes a new method that utilizes the Curvelet transform to remove the resonance interference in original data and the steps are as follows.First,analyze the differences between the characteristics of the resonance interference induced by electrical poles and effective information in the Curvelet domain.Based on this,conduct wavefield separation according to the multi-scale and multi-direction characteristics of the Curvelet transform.Then,further attenuate the interference factors using the nonlinear threshold function designed in this paper.According to the application and analysis of actual data,this method can effectively remove the resonance interference induced by electrical poles while properly protecting effective signals and can significantly improve the signal-to-noise ratio and resolution of the denoised data.Therefore,the method proposed in this paper is effective.