基于Contourlet域自适应Wiener阈值的同时震源波场分离

同时震源数据包含了多炮之间的串扰噪声,不能直接用于常规数据处理流程.因此,需要对混叠的波场进行分离得到常规采集的单炮记录.本文基于稀疏迭代反演分离,提出了一种具有尺度与空间自适应的Wiener阈值选取方法.该阈值选取方法能够根据不同迭代环境计算不同尺度下串扰噪声的方差和不同空间位置有效信号的方差,从而自适应调整阈值大小,最终通过对变换域系数进行收缩来达到去除串扰噪声的目的.理论模型数据和实际数据测试结果表明,本文方法能够快速有效地压制串扰噪声和保护弱有效信号,取得了比Contourlet域子带一致Wiener阈值方法和Curvelet域指数衰减阈值方法更好的分离效果.

Simultaneous-source wavefield separation based on adaptive Wiener threshold in Contourlet domain

The simultaneous-source data cannot be directly used for seismic data processing workflow due to crosstalk noise of the interference guns. Therefore, the blended wavefield needs to be separated to obtain the conventional single shot record. Based on the threshold iterative inversion separation in sparse domain, a scale and spatially adaptive Wiener threshold method is proposed. With the method, the threshold values are adaptively calculated by using the variance of crosstalk noise at different scales and the variance of effective signals at different spatial positions at each iteration, and are used to attenuate the crosstalk noise by shrinking the coefficients in transform domain. The results tested by synthetic and field data show that the proposed method can more quickly and effectively suppress crosstalk noises and keep valuable weak signals than the traditional Wiener threshold method with subband consistent in Contourlet domain and the exponential attenuation threshold method in Curvelet domain.