摘要
本文从地震道的奇异属性出发,利用连续小波变换求取地震道的小波变换模极大值曲线,并沿此曲线提取地震道的小渡变换系数,称之为小波变换模极大值连线振幅。此属性不仅可以代表信号本身,而且可以最大程度的区别于相邻道,并且具有地震道多尺度的特征,即兼具时频域的特征,由此结合自组织神经网络,我们提出了一种新的地震相分析方法,通过模型合成地震记录实验分析,证明此方法是可行的,且对地质层位的解释误差具有一定的容许度,最后,将此方法应用于了实际资料,取得了良好的效果。
This paper discusses the singularity of seismic trace. WTMMLA （wavelet transform modulus maxima line amplitude） can been extracted along WTMMLs （wavelet transform modulus maxima lines） of seismic traces by virtue of wavelet transform. This attribute embodies the essential characteristics of seismic signal in the time-frequency domain and can be used to differentiate from two adjacent seismic traces easily. Therefore, based on the singularity, a new method of seismic faceis analysis is proposed integrated with Kohenon net. This technique has been tested on the theoretical model and then successfully applied to the practical data. This approach has not only succeeded in seismic facies analysis, but also has some ability of tolerance with the horizon interpretation.
出处
《地球物理学进展》
CSCD
北大核心
2009年第6期2220-2225,共6页
Progress in Geophysics
关键词
小波变换
模极大值曲线
奇异属性
神经网络
wavelet transform, modulus maxima Lines, singularity, neural net
