摘要
提出了一种基于域生长算法的初至自动拾取方法。首先是由炮记录中的地震道计算其局部标准方差以及局部互相关等局部相似属性,从而形成局部相似属性道。然后将这些局部属性道作为域生长算法的输入,由区域生长算法根据事先设定的阈值和种子点自动识别出噪声区域和有效信号区域,两个区域的边界即为初至时刻。实例表明,该方法具有一定的抗噪性能,易于使用。
This paper presents a new method to pick the seismic first arrival automatically.First,local attributes such as local standard deviation and local cross-correlation are computed from one trace of shot record.The local attributes are calculated within moving windows along the seismic trace,so that local attributes traces are produced.Then region growing method is used to identify the valid signal region and noise signal region based on the local attributes traces.The location of the edge of the identified two regions can be considered as the first breaking time of the seismic trace.The method has been tested on real data sets,and the results show that,with an appropriate tuning of its parameters,the method is robust under noisy conditions,computationally efficient,and easy to apply.
作者
刘百红
邵志东
郑四连
宋志翔
LIU Bai-Hong;SHAO Zhi-Dong;ZHENG Si-Lian;SONG Zhi-Xiang(Sinopec Geophysical Research Institute,Nanjing 211103,China)
出处
《物探与化探》
CAS
北大核心
2019年第4期843-850,共8页
Geophysical and Geochemical Exploration
基金
国家科技重大专项课题“大中型油气田及煤层气开发”(2016ZX05018007)
关键词
地震勘探
初至
局部属性
自动拾取
区域生长
seismic exploration
first-break
local attributes
automatic pick
region growing method