针对常规方法不能很好地压制复杂地质条件下的低信噪比地震记录中的随机噪声的问题,提出了Contourlet域相关和阈值衰减联合去噪方法(CTCT)。首先,采用Contourlet分解地震资料,对得到的尺度-方向分量进行倾角扫描,通过相关叠加压制随机噪声;然后,对相干加强处理后的剩余随机噪声采用非线性阈值进一步衰减。模型算例及实例结果表明,此方法压制随机干扰的效果要优于传统方法,一定程度上改善了常规滤波方法在压制噪声时对有效波的影响。 Aiming at the problem that the conventional method can not suppress the random noise in low S / N seismograms well under complex geological conditions, the Contourlet domain correlation and thresholding joint denoising method (CTCT) is proposed. Firstly, Contourlet is used to decompose the seismic data, and the obtained scale-directional components are scanned obliquely, and the random noise is suppressed by the correlation superposition. Then, the residual random noise after coherent enhancement processing is further attenuated by nonlinear threshold. The results of numerical examples and examples show that this method is superior to the traditional method in suppressing stochastic disturbances and improves the effective wave effect of the conventional filtering method to a certain extent when suppressing noise.