基于连续小波变换的面波衰减方法研究

Attenuation of ground roll based on continuous wavelet transform

面波压制是陆上地震资料处理的主要任务之一。根据面波和有效波在小波域能量和分布区域的不同,提出利用连续小波变换自动压制面波的方法。首先利用大尺度的小波系数自动识别面波区,然后计算每道面波的频带范围,最后对面波区小波系数进行滤波处理并估计出有效信号,根据计算的面波频带范围对估计出的有效信号进行高通滤波得到去除面波后的地震记录。实际地震资料的处理结果表明,与常用的高通滤波方法相比,该方法自动化程度高,压制面波的同时可以减少对有效信号的影响。

Ground roll attenuation is one of the main tasks of land seismic data processing.According to the energy and distribution characteristics of ground roll and effective waves in wavelet domain,a method is presented to attenuate ground roll using continuous wavelet transform.Firstly,wavelet coefficients of large scale were used to recognize the ground roll region automatically.Then,the ground roll frequency band of each trace was calculated.Finally,the ground roll wavelet coefficients of ground roll region were filtered and the effective signal was approximately obtained.Moreover,the effective signal was filtered using the precalculated ground roll frequency band.Real data examples show that our method has high-degree automation and can attenuate the ground roll effectively with less effective signal loss compared with conventional high-pass filtering methods.