基于信号子空间分解的三维地震资料高分辨率处理方法  

A high-resolution processing technique for 3-D seismic data based on signal sub-space decomposition

作　　者：陆文凯[1] 丁文龙[1] 张善文[2] 肖焕钦[2] 赵铭海[2] 

机构地区：[1]清华大学自动化系信息处理所,智能技术和系统国家重点实验室,北京100084 [2]胜利油田有限公司,山东东营257100

出　　处：《地球物理学报》2005年第4期896-901,共6页Chinese Journal of Geophysics

基　　金：国家科技攻关(2003BA613A02);国家自然科学基金(40474040);中国石油天然气集团公司创新基金资助.

摘　　要：提出了一种新的基于信号子空间分解技术的3D地震资料高分辨率处理方法.利用信号子空间分解技术,不仅可以分离3D地震资料中的信号和噪声,而且可以进一步根据地震同相轴的倾角不同,将混合信号进行分解得到单个同相轴信号.通过丢弃噪声子空间,只对不同信号子空间重构的信号利用谱白化技术进行高分辨率处理,然后累加所有处理结果,从而达到既提高地震资料的分辨率,又提高地震资料信噪比的目的.对合成资料和实际地震资料的处理结果表明,此法具有好的应用前景.This paper proposes a new signal sub-space decomposition(SSD) based technique to improve the resolution of 3-D seismic data processing. By exploiting the SSD technique, we can decompose not only the signal and noise in 3-D seismic data, but also the composite event into single events according to their different dips. After removal of the noise space, we integrate all the reconstructions corresponding to the signal sub-spaces with improved resolution using the spectral whitening technique. Since we only improve the resolution of the signal and remove the noise at the same time, we can improve the resolution and the signal_to_noise ratio (SNR) of the seismic data simultaneously. The application results of the synthetic and real data show our method has a good perspective of applications.

关 键 词：信号子空间分解 奇异值分解 谱白化 分辨率 信噪比 

分 类 号：P631[天文地球—地质矿产勘探]