基于块匹配协同滤波的三维地震信号去噪

Three-dimensional seismic signal denoising based on block matching and collaborative filtering

提出了一种基于块匹配协同滤波的三维地震信号去噪算法.该算法将块匹配四维协同滤波(BM4D)算法基础估计中的小波变换用离散余弦变换取代,提高了BM4D基础估计的信噪比,同时给BM4D最终估计中的经验维纳滤波提供更接近原始信号的能量谱,从而达到提高整体去噪效果的目的.合成三维地震信号和实际复杂三维地震信号去噪结果表明,和基础估计分别采用bior1.5、bior2.4、dmey小波变换的BM4D算法相比,该算法对三维地震信号具有更优的去噪效果.

A three-dimensional seismic signal denoising algorithm based on block-matching and collaborative filtering is proposed. This algorithm replaces the wavelet transform in the basic estimation of block matching 4D filtering （BM4D） al- gorithm with discrete cosine transform, which has improved the signal to noise ratio of BM4D basic estimates. Give a closer energy spectrum of the original signal to experience wiener filtering of BM4D final estimate so as to improve the overall de- noising effect. The denoising results on synthetic three-dimensional seismic signals and real complex three-dimensional seismic signals illustrate that the proposed method can denoise the three -dimensional seismic signals very well, compar- ing with BM4D denoising algorithm basic estimates based on bior1.5, bior2.4 and dmey wavelet transform.