利用Shearlet变换提高叠后地震资料分辨率  

Improving post-stack seismic data resolution based on Shearlet transform

作　　者：郭爱华 路鹏飞 余波[4] 卢成慧 汪彬[5] 万玲娜 GUO Aihua;LU Pengfei;YU Bo;LU Chenghui;WANG Bin;WAN Lingna(School of Information Engineering,East China University of Technology,Nanchang,Jiangxi 330013,China;Jiangxi Engineering Technology Research Center of Nuclear Geoscience Data Science and System,East China University of Technology,Nanchang,Jiangxi 330013,China;Jiangxi Engineering Laboratory on Radioactive Geoscience and Big Data Technology,East China University of Technology,Nanchang,Jiangxi 330013,China;Baikouquan Oil Production Plant of Petrochina Xinjiang Oilfield Branch,Karamay,Xinjiang 834000,China;School of Software,East China University of Technology,Nanchang,Jiangxi 330013,China)

机构地区：[1]东华理工大学信息工程学院,江西南昌330013 [2]东华理工大学江西省核地学数据科学与系统工程技术研究中心,江西南昌330013 [3]东华理工大学江西省放射性地学大数据技术工程实验室,江西南昌330013 [4]中国石油新疆油田分公司百口泉采油厂,新疆克拉玛依834000 [5]东华理工大学软件学院,江西南昌330013

出　　处：《石油地球物理勘探》2021年第5期992-1000,I0009,共10页Oil Geophysical Prospecting

基　　金：江西省核地学数据科学与系统工程技术研究中心开放基金项目“地震大数据高分辨率处理方法研究”(JETRCNGDSS202106)资助。

摘　　要：反褶积、Q补偿、谱白化、小波变换等方法在提高地震资料分辨率的同时往往会放大噪声,降低地震资料的信噪比。由于地震随机噪声服从高斯分布且其本身并无方向性,因而在Shearlet域可将有效信号与随机噪声分开。通过Shearlet变换将地震信号转换到Shearlet域,对Shearlet域系数进行合理的补偿后,再做Shearlet反变换,可实现对地震资料的提高分辨率处理。结合Shearlet变换的这两个特点,首先舍弃随机噪声Shearlet域系数,同时只对优势频带范围内的Shearlet域系数做补偿和提频处理。这样既提高地震资料分辨率又保持了地震资料信噪比。合成地震数据和叠后实际资料的处理结果表明,本文方法可有效提高叠后地震资料分辨率。Deconvolution,Q-compensation,spectral whitening,wavelet transform,and other methods often enlarge noise and reduce the signal-to-noise ratio of seismic data while improving the resolution of seismic data.Since seismic random noise obeys Gaussian distribution and has no directivity,effective signals and random noise can be separated in the Shearlet domain.The seismic signal is transformed into the Shearlet domain by Shearlet transform.The coefficients in the Shearlet domain are compensated reasonably,and then the inverse Shearlet transform is carried out,which can improve the resolution of seismic data.Combined with the two characteristics of Shearlet transform,firstly,the random noise coefficients in the Shearlet domain are discarded,and at the same time,only the coefficients in the Shearlet domain within the dominant frequency band are compensated,and the frequency is raised.This not only improves the resolution of seismic data but also maintains the signal-to-noise ratio of seismic data.The proces-sing results of synthetic seismic data and post-stack real data show that this method can improve the resolution of post-stack seismic data.

关 键 词：叠后地震资料 SHEARLET变换 能量补偿 分辨率 信噪比 

分 类 号：P631[天文地球—地质矿产勘探]