基于Curvelet变换的缺失地震数据插值方法

被引:38
作者
刘国昌
陈小宏
郭志峰
刘华锋
高建军
机构
[1] 中国石油大学(北京)油气资源与探测国家重点实验室
关键词
地震数据; 重建; Curvelet变换; 凸集投影; L2范数; L1范数;
D O I
10.13810/j.cnki.issn.1000-7210.2011.02.014
中图分类号
P631.44 [];
学科分类号
摘要
Curvelet变换具有局部性、多尺度性和多方向性,在处理非稳态地震信号中具有较大的优势,可以利用Curvelet变换的压缩特性重建缺失的地震数据。本文首先分析了基于稀疏变换的缺失地震数据插值的基本原理,在反问题正则化理论框架下,针对L2范数约束和L1范数约束条件,分析了两种约束的差异,着重阐述了基于Curvelet变换的L1范数约束的插值方法,其优点在于对非线性同相轴不需要分窗口处理,并将凸集投影算法(POCS)引入到Curvelet变换的插值方法中,通过采用按指数规律衰减的阈值参数加快了迭代收敛的速度。理论和实际算例验证了Curvelet变换插值方法的有效性。
引用
收藏
页码:237 / 246+159 +159
页数:11
相关论文
共 40 条
[11]  
De-noising by soft-thresholding. David L Donoho. IEEE Transactions on Information Theory . 1995
[12]  
New tight frames of curvelets and optimal representations of objects with C~2 singularities. Candes E J,Donoho D L. Communications of the ACM . 2004
[13]  
Leading-order seismic imaging using curvelets. Douma,H,de Hoop,M. Geophysics . 2007
[14]  
3D interpolation of irregular data with a POCS algorithm. R. Abma,N. Kabir. Geophysics . 2006
[15]  
Practical signal recovery from random projections. Candes E J,Romberg J. Proceedings of SPIE Conference Wavelet Applications in Signal and Image ProcessingⅪ . 2005
[16]  
Interpolation with smoothly nonstationary prediction-error filters. Crawley S,Clapp R,Claerbout J. 69th Ann.Internat.Mtg,Soc.Expl.Geophys.Expanded abstracts . 1999
[17]  
Basic Earth I maging:Stanford Explo-ration Project. Claerbout J F. http:∥sepwww.stanford.edu/sep/prof/bei1005.pdf . 1995
[18]  
Reverse ti me migration with opti malcheckpointing. Symes W W. Technical Report 06-18,Depart mentof Computational and Applied Mathematics.Rice U-niversity,Houston,Texas,USA . 2006
[19]  
Stacking angle-domaincommon-i mage gathers for normalization of illumina-tion. Liu G C,Fomel S,Chen X. Geophysical Prospecting . 2010
[20]  
Parametric Reconstruction of Multi-di mensional Seismic Records. Naghizadeh M. . 2009