众所周知,地震记录通常被描述为地震子波与反射系数序列的褶积。我们所说的反褶积处理就是要从地震记录中求取反射系数序列。它从图形上表示为一列脉冲,如能找到这些峰值就等于找到了它们的振幅和发生的时间,也等于提供了地下界面的几何和地质的信息。最常规的反褶积法是基于维纳滤波的理论。一般它要求时不变和平稳的条 It is well known that seismic records are usually described as the convolution of seismic wavelet and reflection coefficient sequence. What we mean by deconvolution is to get the reflection coefficient sequence from the seismic record. It is graphically represented as a series of pulses, which, if found, equals the amplitude and timing at which they were found, as well as providing geometrical and geological information about the subsurface. The most common deconvolution method is based on Wiener filtering theory. Generally it requires constant and stable strips