基于高阶相关的Curvelet域和空间域的倾角扫描噪声压制方法
|
摘要
低信噪比地震数据处理一直是人们关注的问题。本文基于Curvelet多尺度多角度分析理论及高阶统计理论,提出高阶相关的Curvelet域和空间域倾角扫描压制随机噪声方法,并从向量量化角度解释了该方法的基本原理。文中分析了Curvelet域θ角度特征,提出利用θ作为倾角扫描的角度因子并进行空间域高阶相关计算,实现了对复杂区地震数据的压噪处理。通过对设定模型算例及实际资料的处理计算,验证了该方法在保持有效反射波信息的前提下,可有效压制随机干扰,所得地震剖面的信噪比比常规滤波处理方法有显著的提高。
Seismic data with low Signal to Noise Ratio is always a noticeable point, based on multi-scale and multi-angle analysis theory and high order statistic theory, dip angle scanning noise elimination method in Curvelet domain and space domain was presented in this paper, the basic principle of this method was explained from the point of view in vector quantization. In this paper θ angle characteristics in Curvelt domain was analyzed, and it is suggested that θ was used as angle factor in dip angle scanning and then the high order correlation calculation in space domain was conducted, as the result the noise elimination for seismic data in complex area can be realized. Processing results for model data and field data show that random interference were effectively suppressed while effective reflection signal can be kept, the Signal-to-Noise Ratio for seismic section by this method is greatly raised compared with routine filtering processing methods.
Seismic data with low Signal to Noise Ratio is always a noticeable point, based on multi-scale and multi-angle analysis theory and high order statistic theory, dip angle scanning noise elimination method in Curvelet domain and space domain was presented in this paper, the basic principle of this method was explained from the point of view in vector quantization. In this paper θ angle characteristics in Curvelt domain was analyzed, and it is suggested that θ was used as angle factor in dip angle scanning and then the high order correlation calculation in space domain was conducted, as the result the noise elimination for seismic data in complex area can be realized. Processing results for model data and field data show that random interference were effectively suppressed while effective reflection signal can be kept, the Signal-to-Noise Ratio for seismic section by this method is greatly raised compared with routine filtering processing methods.
引文
[1]李庆忠.走向精确勘探的道路——高分辨地震勘探系统工程剖析.北京:石油工业出版社,1993
[2]高静怀,毛剑,满蔚仕等.叠前地震资料噪声衰减的小波域方法研究.地球物理学报,2006,49(4):1155~1163
[3]勾丽敏,蔡希玲,刘学伟等.噪声对叠前深度偏移层速度精度的影响分析.石油地球物理勘探,2007,42(2):156~163
[4]Spitzer R,Nitsche F O,Green A G.Reducing sourcegenerated noise in shallow seismic data using linear and hyperbolic tau-p transformations.Geophysics,2001,66(5):1612~1621
[5]Kelamis P G,Verschuur DJ.Surface-related multiple eli mination on land seismic data-strategies via case studies.Geophysics,2000,65(3):719~734
[6]Benyamin N.Key elements of total seismic field de-sign using mathematica-a tutorial.Geophysics,2002,67(4):1020~1027
[7]Wang Y H.Quantifying the effectiveness of stabilized inverse Q filtering.Geophysics,2003,68(1):337~345
[8]Wang Y H.Inverse Q-filter for seismic resolution en-hancement.Geophysics,2006,71(1):51~60
[9]Mallat S.Atheory for multiresolution signal decom-position:The wavelet representation.IEEE Transac-tions on Pattern Analysis and Machine Intelligence,1989,11(7):674~693
[10]张三宗,徐义贤.地震记录小波域高阶相关叠加技术.地球物理学报,2006,49(2):554~560
[11]Cands E J,Donoho D L.Ridgelets:a key to higher-di mensional intermittency.Phil Trans Roy Soc Lon-don A,1999,357(1760):2495~2509
[12]Donoho D L.Ridge functions and orthonormal ridge-lets.Journal of Approxi mation Theory,2001,111:143~179
[13]Arivazhagan S,Ganesan L and Subash Kumar T G.Texture classification using ridgelet transform.Pat-tern Recognition Letters,2006,27:1875~1883
[14]Zhang H L,Liu T Y.Study of Ridgelet transformin i mproving S/Nratio of seismic data in South China.Journal of China University of Geosciences,2007,18:507~509
[15]Jean-Lue Starck,Cands E J and Donoho D L.The curvelet transformfor i mage denoising.IEEE Trans on Image Processing,2002,11(6):670~684
[16]Cands E J,Donoho D L.Continuous curvelet trans-form.Appl Comput Harmon Anal,2005,19:198~222
[17]王书明,朱培民,李宏伟等.地球物理学中的高阶统计量方法.北京:科学出版社,2006
[18]唐斌,尹成.基于高阶统计的非最小相位地震子波恢复.地球物理学报,2001,44(3):404~410
[19]熊小军,尹成,张白林等.高阶统计量油气检测方法研究.地球物理学报,2004,47(5):920~927
[20]王书明,王家映.高阶统计量在大地电磁测深数据处理中的应用研究.地球物理学报,2004,47(5):928~934
[21]An Y,Wei L C,Yang C C.The most homogeneous dip-scanning method using edge preserving smoothing for seismic noise attenuation.Apply Geophysics,2006,3(4):210~217
[22]夏洪瑞,周开明.地震资料处理中相干干扰消除方法分析.石油物探,2003,42(4):526~528
[2]高静怀,毛剑,满蔚仕等.叠前地震资料噪声衰减的小波域方法研究.地球物理学报,2006,49(4):1155~1163
[3]勾丽敏,蔡希玲,刘学伟等.噪声对叠前深度偏移层速度精度的影响分析.石油地球物理勘探,2007,42(2):156~163
[4]Spitzer R,Nitsche F O,Green A G.Reducing sourcegenerated noise in shallow seismic data using linear and hyperbolic tau-p transformations.Geophysics,2001,66(5):1612~1621
[5]Kelamis P G,Verschuur DJ.Surface-related multiple eli mination on land seismic data-strategies via case studies.Geophysics,2000,65(3):719~734
[6]Benyamin N.Key elements of total seismic field de-sign using mathematica-a tutorial.Geophysics,2002,67(4):1020~1027
[7]Wang Y H.Quantifying the effectiveness of stabilized inverse Q filtering.Geophysics,2003,68(1):337~345
[8]Wang Y H.Inverse Q-filter for seismic resolution en-hancement.Geophysics,2006,71(1):51~60
[9]Mallat S.Atheory for multiresolution signal decom-position:The wavelet representation.IEEE Transac-tions on Pattern Analysis and Machine Intelligence,1989,11(7):674~693
[10]张三宗,徐义贤.地震记录小波域高阶相关叠加技术.地球物理学报,2006,49(2):554~560
[11]Cands E J,Donoho D L.Ridgelets:a key to higher-di mensional intermittency.Phil Trans Roy Soc Lon-don A,1999,357(1760):2495~2509
[12]Donoho D L.Ridge functions and orthonormal ridge-lets.Journal of Approxi mation Theory,2001,111:143~179
[13]Arivazhagan S,Ganesan L and Subash Kumar T G.Texture classification using ridgelet transform.Pat-tern Recognition Letters,2006,27:1875~1883
[14]Zhang H L,Liu T Y.Study of Ridgelet transformin i mproving S/Nratio of seismic data in South China.Journal of China University of Geosciences,2007,18:507~509
[15]Jean-Lue Starck,Cands E J and Donoho D L.The curvelet transformfor i mage denoising.IEEE Trans on Image Processing,2002,11(6):670~684
[16]Cands E J,Donoho D L.Continuous curvelet trans-form.Appl Comput Harmon Anal,2005,19:198~222
[17]王书明,朱培民,李宏伟等.地球物理学中的高阶统计量方法.北京:科学出版社,2006
[18]唐斌,尹成.基于高阶统计的非最小相位地震子波恢复.地球物理学报,2001,44(3):404~410
[19]熊小军,尹成,张白林等.高阶统计量油气检测方法研究.地球物理学报,2004,47(5):920~927
[20]王书明,王家映.高阶统计量在大地电磁测深数据处理中的应用研究.地球物理学报,2004,47(5):928~934
[21]An Y,Wei L C,Yang C C.The most homogeneous dip-scanning method using edge preserving smoothing for seismic noise attenuation.Apply Geophysics,2006,3(4):210~217
[22]夏洪瑞,周开明.地震资料处理中相干干扰消除方法分析.石油物探,2003,42(4):526~528
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心