三维曲波变换在地震资料去噪处理中的应用研究

考虑到二维曲波变换对于三维地震数据难以达到理想的去噪效果,提出了一种基于三维曲波变换的地震资料去噪方法。通过三维曲波变换使三维地震数据变换到曲波域,将三维地震数据分为多个方向和尺度,利用相关计算法判别代表有效信号和随机噪声的曲波系数,并根据曲波变换的性质,采用改进的非线性阈值方法对曲波系数进行处理,最后通过三维曲波反变换得到去噪后的地震信号。模型数据及实际地震资料的处理结果表明,三维曲波变换去噪处理方法可以有效压制三维地震资料中的随机噪声,同时较好地保护有效信号,提高了地震资料的信噪比。

Owing to the undesired effect of 3D seismic data denoising based on 2D curvelet transform,a seismic data denoising method based on 3D curvelet transform is proposed in this paper.First,3D seismic data is transformed to curvelet domain,which is decomposed to different scales and different orientations.And then we make correlation calculation to distinguish curvelet coefficients representing signal or random noise,use modified non-linear threshold to process the curvelet

coefficients.Finally,denoised seismic signal is obtained by inverse 3D curvelet transform.The test processing results of modeling and actual data show that the denoising method based on 3D curvelet transform could suppress random noise and preserve effective signals as to improve the SNR of the 3D seismic data.