基于时频变换的低信噪比地震信号初至拾取

初至拾取是勘探地震资料处理中最基础的工作之一,现有的初至拾取方法日趋成熟.但当信号信噪比较低时,常规方法的拾取精度随之降低,因此一些适用于低信噪比数据的拾取方法被提出.其中应用较广泛的是基于时频分析的初至拾取方法,由于拾取过程涉及时频正变换和逆变换,效率较低,鉴于此,基于时频系数的叠加结果和时间域信号具有相同波形特征的特点,提出了一种在时频域直接开展初至拾取的方法,首先将数据转入时频域,然后采用自适应的噪声衰减方法进行噪声压制,最后在时频域直接拾取初至.模拟和实际数据的拾取结果证实,该方法在低信噪比的环境下,可以获得高于常规方法的拾取精度,并且比基于小波变换的其他方法更节省计算成本.

First brick pick of low SNR seismic signal based on time-frequency transformation

The first brick picking is one of the most basic works in the processing of seismic data,and the existing picking method is becoming more mature.However,when the SNR of seismic signal is low,the picking accuracy of the conventional method decreases.Therefore some pickup methods suitable for low SNR data are proposed,among which the first brick picking method based on time-frequency analysis is more widely used.The picking process involves time-frequency forward transformation and inverse transformation,hence the efficiency is low.In view of this,on the basis that the superposition results of the time-frequency coefficient and the signals in time domain have the same waveform characteristics,a direct picking method in time-frequency domain is proposed.First the data was transformed into the time-frequency domain,and then the adaptive noise attenuation method was used for noise suppression.Finally,pick the first brick directly in the time-frequency domain.The picking results of simulation and field data show that the method can obtain higher picking accuracy than the conventional method in the environment of low signal-to-noise ratio,and save more cost of computing than other methods based on wavelet transform.