小波变换回波提取在可控震源地震信号处理中的应用

在传统的可控震源地震信号处理中,通常用相关算法和反褶积算法把各地震道采集的原始数据转换成地震剖面,然而它们的效果受环境噪声的干扰明显。本文采用一种基于小波变换的时频互相关(TFCC)算法,能从可控震源地震数据中检测回波并估计其时间延迟。在该算法中,首先对道数据和扫频源信号进行小波变换,得到其时频表示,然后对其时频表示进行互相关。在时频互相关的计算结果中,回波被转变成时频相关子波,同时原始道数据被转变成地层剖面。用TFCC算法和相关算法来处理相同的实际可控震源地震数据,显示前者能更好地压制噪声,并能清晰地检测到微弱回波信号。

Application of wavelet transform to pick up reflection wave in vibroseis seismic data processing

In traditional vibroseis' seismic data processing, the correlation and deconvolution algorithms are usually used to transform acquired raw data in separate seismic channels into seismic sections, but their effects are severely affected by the interference of environment noises. The paper adopts the time-frequency cross-correlation (TFCC) algorithm based on wavelet transform that can detect the reflection waves from vibroseis seismic data and estimate time delays. In the algorithm, firstly the wavelet transform is carried out for channel data and sweeping signals, which results in time-frequency representation; then, the cross-correlation is carried out for time-frequency representation. The reflection waves are transformed into time-frequency correlated wavelet and raw channel data is transformed into formation section at the same time in computed results of time-frequency cross-correlation. Using TFCC and correlation algorithms for processing of same practical vibroseis seismic data showed that the former could better suppress noises and detect weak reflection signal distinctively.