基于Curvelet域高斯尺度混合模型的地震信号降噪方法

摘要

结合曲波变换和高斯尺度混合模型提出地震信号随机噪声压制方法.该方法首先运用曲波变换对含有随机噪声的地震信号进行分解,然后对各小波子带系数分别建立高斯尺度混合模型估计出原始地震信号所对应的小波系数,最后经曲波逆变换重构获得降噪处理后的地震信号.仿真地震信号和实际地震信号的实验结果均表明本文方法能够有效压制地震信号中的随机噪声干扰,较多地保留了有效信号.%The method of random noise suppression of seismic data is proposed combining Curvelet transform with Gaussian scale mixture model.Firstly, The seismic signal containing random noise is decomposed by Curvelet transform.Then the gaussian scale mixture models are constructed respectively for each wavelet sub-band coefficients to estimate the source seismic signal wavelet coefficients.Finally, the denoised seismic signal can be reconstructed using the Curvelet inverse transform.Through experiments on synthetic records and the actual seismic signal, the results demonstrate that the proposed method can attenuate random noise of the seismic signal effectively and keep more useful signal.