Ridgelet变换在地震数据压缩中的应用
|
摘要
鉴于地震数据反映的物理界面的空间展布往往具有直线、平面等特性,小波变换进行震压缩处理不能很好地反映这一特性,根据Ridgelet变换的特点,将Ridgelet变换应用于地震数据压缩,结合嵌入式零树编码方法,提出了Ridgelet变换的地震数据压缩方法。通过对实际资料的处理,在压缩率为99.0%和90%时,比较了Ridgelet变换与小波变换处理结果。研究结果表明:Ridgelet变换应用于地震数据压缩,其对数据的压缩比比小波变换对数据的压缩比大。
Wavelet transform has been introduced successfully to seismic data compression, it is efficient to the objects of point singularities according to its characters. In seismic data imaging, because the layers media is of line (or super-plane) singularities, wavelet transform has its shortcomings. Recently, a new transform, i.e., Ridgelet transform, was proposed. Ridgelet transform has much higher precision, especially for describing the objects which have line (or super-plane) singularities. In this paper, according to this character of ridgelet transform, the method of seismic data compressing, used Ridgelet transform is proposed. The results of the method are obtained for real seismic data. Practical experiments results show that the proposed method works well. With the comparisons to the wavelet transform, at the compression ratio 99.0% and 90%, the Ridgelet transform is more efficient and has higher compressing ratio, and is much better than other methods.
Wavelet transform has been introduced successfully to seismic data compression, it is efficient to the objects of point singularities according to its characters. In seismic data imaging, because the layers media is of line (or super-plane) singularities, wavelet transform has its shortcomings. Recently, a new transform, i.e., Ridgelet transform, was proposed. Ridgelet transform has much higher precision, especially for describing the objects which have line (or super-plane) singularities. In this paper, according to this character of ridgelet transform, the method of seismic data compressing, used Ridgelet transform is proposed. The results of the method are obtained for real seismic data. Practical experiments results show that the proposed method works well. With the comparisons to the wavelet transform, at the compression ratio 99.0% and 90%, the Ridgelet transform is more efficient and has higher compressing ratio, and is much better than other methods.
引文
[1] 冯占林,张学工,李衍达.基于小波变换的地震勘探数据压缩的工程分析[J].清华大学学报(自然科学版),2001,41(5):170173.FENGZhan lin,ZHANGXue gong,LIYan da.Somepracticalaspectsofseismicdatacompressionbasedonwavelet[J].JTsinghuaUni(Sci＆Tech),2001,41(5):170173.
[2] 赵改善.小波变换与地震数据压缩[J].石油物探,1994,33(4):1927.ZHAOGai shan.Wavelettransformandseismicdatacompression[J].GeophysicalProspectingforPetroleum,1994,33(4):1927.
[3] 唐向宏,贺振华,杨绍国.利用小波变换进行地震数据压缩[J].成都理工学院学报,1999,26(2):183186.TANGXing hong,HEZhen hua,YANGShao guo.SeismicdatacompressionusingM bandwavelettransform[J].JofChenduUnivofTechnology,1999,26(2):183186.
[4] CandesEJ.Ridgelet:theoryandapplications[D].Stanford:DepartmentofStatistics,StanfordUniversity,1998.
[5] CandesEJ.Ridgelets:Estimatingwithridgefunctions[R].Stanford:DepartmentofStatistics,StanfordUniversity,1999.
[6] CandesEJ.Monoscaleridgeletsfortherepresentationofimageswithedges[R].Stanford:DepartmentofStatistics,StanfordUniversity,1999.
[7] DonohoDL.OrthonormalRidgeletsandlinearsingularities[J].SIAMJMathAnal,2000,31(5):10621099.
[8] DonohoDL.DigitalRidgelettransformviarectopolarcoordinatetransform[R].Stanford:DepartmentofStatistics,StanfordUniversity,1998.
[9] BeylkinG.DiscreteRadontransform[J].IEEETansASSP,1987,35(1):162172.
[10] DonohoDL.De noiseingbysoft thresholding[J].IEEETransInfomTheory,1995,41(5):613627.
[11] DonohoDL,JohnstoneIM.Idealspatialadaptationviawaveletshrinkage[J].Biometrika,1994,81:425455.
[12] ShapiroJM.Embeddedimagecodingusingzerotreeofwaveletcoefficients[J].IEEETransSignalProcessing,1993,41(12):34453462.
[13] 侯 颖,李国民.一种嵌入式小波零树的图像分割编码算法[J].西安科技学院学报,2003,23(1):7477.HOUYing,LIGuo min.Imagesegmentationcodingalgorithmbasedonembeddedwaveletzero treecodingscheme[J].JofXi'anUnivofScienceandTechnology,2003,23(1):7477.
[14] 郑 勇,周正华,朱维乐.基于多级零树编码的小波系数网格编码量化[J].电子与信息学报,2002,24(12):18991905.ZHENGYong,ZHOUZheng hua,ZHUWei le.TCQofwaveletcoefficientsbasedonmulti stagezerotreecoding[J].JofElectronicsandInformTech,2002,24(12):18991905.
[2] 赵改善.小波变换与地震数据压缩[J].石油物探,1994,33(4):1927.ZHAOGai shan.Wavelettransformandseismicdatacompression[J].GeophysicalProspectingforPetroleum,1994,33(4):1927.
[3] 唐向宏,贺振华,杨绍国.利用小波变换进行地震数据压缩[J].成都理工学院学报,1999,26(2):183186.TANGXing hong,HEZhen hua,YANGShao guo.SeismicdatacompressionusingM bandwavelettransform[J].JofChenduUnivofTechnology,1999,26(2):183186.
[4] CandesEJ.Ridgelet:theoryandapplications[D].Stanford:DepartmentofStatistics,StanfordUniversity,1998.
[5] CandesEJ.Ridgelets:Estimatingwithridgefunctions[R].Stanford:DepartmentofStatistics,StanfordUniversity,1999.
[6] CandesEJ.Monoscaleridgeletsfortherepresentationofimageswithedges[R].Stanford:DepartmentofStatistics,StanfordUniversity,1999.
[7] DonohoDL.OrthonormalRidgeletsandlinearsingularities[J].SIAMJMathAnal,2000,31(5):10621099.
[8] DonohoDL.DigitalRidgelettransformviarectopolarcoordinatetransform[R].Stanford:DepartmentofStatistics,StanfordUniversity,1998.
[9] BeylkinG.DiscreteRadontransform[J].IEEETansASSP,1987,35(1):162172.
[10] DonohoDL.De noiseingbysoft thresholding[J].IEEETransInfomTheory,1995,41(5):613627.
[11] DonohoDL,JohnstoneIM.Idealspatialadaptationviawaveletshrinkage[J].Biometrika,1994,81:425455.
[12] ShapiroJM.Embeddedimagecodingusingzerotreeofwaveletcoefficients[J].IEEETransSignalProcessing,1993,41(12):34453462.
[13] 侯 颖,李国民.一种嵌入式小波零树的图像分割编码算法[J].西安科技学院学报,2003,23(1):7477.HOUYing,LIGuo min.Imagesegmentationcodingalgorithmbasedonembeddedwaveletzero treecodingscheme[J].JofXi'anUnivofScienceandTechnology,2003,23(1):7477.
[14] 郑 勇,周正华,朱维乐.基于多级零树编码的小波系数网格编码量化[J].电子与信息学报,2002,24(12):18991905.ZHENGYong,ZHOUZheng hua,ZHUWei le.TCQofwaveletcoefficientsbasedonmulti stagezerotreecoding[J].JofElectronicsandInformTech,2002,24(12):18991905.
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