基于小波变换与奇异值分解的地震资料去噪新方法
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
为了有效地去除地震资料中的随机噪声,充分利用小波变换(WT)去噪和奇异值分解(SVD)去噪方法的优点,提出了一种新的基于小波变换和奇异值分解(WT-SVD)的地震资料去噪方法。该方法首先进行小波软阈值去噪,有效地降低噪声的方差;然后进行基于倾角扫描的奇异值分解去噪,识别噪声点,自动追踪同相轴,并进行同相轴拉平处理,充分利用了奇异值分解方法处理水平同相轴噪声效果好的优点。理论模型和实际资料的去噪结果表明,该研究提出的WT-SVD方法简单易行,比单一的SVD方法和WT方法的去噪效果更显著,有效地消除了地震资料中的随机噪声,显著地提高了地震资料的信噪比。
In order to effectively remove the random noise in seismic data, a new denoising method was proposed by fully using wavelet transform (WT) and singular value decomposition (SVD) . In this method, first "wavelet soft-threshold denoising" could effectively reduce the noise variance, second, SVD based on dip scan could identify noise points, track event automatically ,correct the direction of horizontal event, the advantage of SVD effectiveness was fully used for removing the noise of horizontal event. The results of theoretical model and real data show that the proposed denoising method is easy and feasible, can effectively eliminate the random noise and significantly improve the signal to noise ratio of seismic data.
引文
[1]Stéphane Mallat.信号处理中的小波导引[M].杨力华等译.北京:机械工业出版社,2002.552~558.
    [2]Cleofe M,Ramirez V.Artificial neural networktechnique rainfallforecasting appliedtothe Sao Paulo region[J].Journal of Hydrology,2005,301(1/4):146~162.
    [3]Levent S,Selesnick I W.Bivariate shrinkage functionfor wavelet based denoising exploitinginterscale de2 pendency[J].IEEE Transon Signal Processing,2002,12(50):2744~2756.
    [4]黄中敏,游洪文,孟小红.阿曼3638区块低信噪比地震资料处理技术[J].石油天然气学报,2007,29(1):63~65.
    [5]钟本善,何昌礼,杨忠民.复杂构造地区的SVD去噪技术[J].成都理工学院学报,2000,27(1):93~95.
    [6]黄捍东,张如伟,郭迎春.地震信号的小波分频处理[J].石油天然气学报,2008,30(3):158~159.
    [7]刘鑫,李可恩,贺振华.基于小波变换的地震资料局域自适应去噪研究[J].勘探地球物理进展,2007,30(4):189~191.
    [8]安勇,杨长春.一种改进的频率-波数域倾角扫描去噪方法[J].石油地球物理勘探,2008,43(2):210~212.

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