最小二乘叠前时间偏移在地震数据规则化中的应用

地震数据不规则和地下照明不均匀是阻碍地震保幅成像的两个关键因素。为此，利用已知的地下模型信息，采用Kirchhoff叠前时间偏移将原始数据映射到成像空间，在成像空间提取有效特征，再反偏移到数据空间，重建缺失数据。由于偏移算子在一定程度上消除了地震波的传播效应，因此成像空间比数据空间更加简单，更易于提取有效的信息特征。在最小二乘意义下利用迭代算法拟合观测数据从而提高了方法的精度，并分析了该方法对速度模型的依赖性。理论模型和实际资料处理结果表明，该方法能够对复杂介质条件下的地震数据进行有效的规则化处理，并改善成像质量。

Seismic data regularization via least-squares prestack time migration

Seismic data regularization and uneven subsurface illumination are two of key factors hindering amplitude-preserve imaging.Based on information of known subsurface model,original data are mapped to the image space with Kirchhoff migration operator.Then features are extracted in the image space.Finally the data are demigrated to the data space.Since the migration operator eliminates propagating effects to some extent,it is easier to extract features in the image space.In this paper,we propose an improved seismic regularizationmethod using an iterative algorithm based on least-squares inversion and analyze its dependence on velocity model.Considering the computational cost and flexibility,Kirchhoff migration and demigration are used as the mapping operators.Synthetic and real data examples show that the proposed method is able to reconstruct missing data and improve final images even in complex geological settings.