多尺度地震信号属性分析

MultiScale Sismic Signal Attribute Analysis

地震信号的瞬时属性分析对地震岩性解释和烃类检测很有帮助，通常采用希尔伯特变换的复数道分析方法。利用这种方法，可以比较方便地从地震记录中分离出波的瞬时振幅、瞬时频率、瞬时相位等参数。信号的不同频带具有不同的属性特征，而小波变换具有对信号进行多尺度分解的特性，如采用解析小波则分解后的小波系数仍为一解析信号，且其虚部是其实部的希尔伯特变换。因此根据这一特点可以研究地震信号在不同分辨率下的属性特征。

Seismic signal attribute analysis plays an important role in the interpretation of seismic data. Traditionally, we computed attribute of the complex trace in terms of the Hilbert transform. It is a considerly convinent method. However, the attributes we computed belong to the entire frequency band. The attribute of seismic signal varies with frequency. For the multiresolution character of wavelet transform, we can study the multiscale attribute of seismic data using the complex wavelet coefficients. Thus the imaginary part of wavelet coefficients is the Hilbert transform of the real part.