基于矢量中值滤波的ISS高密度地震数据去噪方法

A deblending method based on vector median filter for ISS high density seismic data

ISS高密度地震采集是以干扰炮相对有效炮为随机信号的基本理念对地震数据进行采集的方法,其大部分有效信号会淹没于噪声之中。基于此,本文提出一种在十字排列域或检波点域进行矢量中值滤波方法,并针对理论模型简要分析了不同空间滤波参数对保持有效信号和压制干扰信号的影响及处理中应注意的问题;实际单炮地震数据的处理和叠加成像结果表明,该方法在有效压制相干炮噪声的同时,能够较好地保持有效信号,提高资料信噪比,显著优于常规去噪方法的处理效果。

As a high efficiency and high density acquisition method,the vibroseis independent simultaneous sweeping(ISS) is different from the other conventional acquisitions.The signals are almost completely covered by multi-coherent shots,which are considered as the random noise in the normal shots.In this paper,we propose a new vector median filter method for deblending the noise in the cross-spread or receiver domains for ISS data.Based on a theory model,we briefly analyze the affections on signal and noise by the different spatial filter parameters,and point out what should be paid attention to during processing.The model and real data results demonstrate that the new method can improve the signal to noise ratio and preserve signals.