改进阈值法在二代小波变换去噪中的应用

地震信号去噪能有效提高信号的信噪比和分辨率。二代小波变换可以在不同尺度上对含噪信号进行小波分解和多分辨率分析,实现窗口宽度自适应调整的局部化分析。但小波变换阈值法在去噪过程中会在信号的不连续邻域会产生伪吉布斯效应,而平移不变量阈值去噪方法通过平移-去噪-平均的思想可以很好的解决该问题。因此本文在已有的二代小波变换阈值去噪的基础上将平移不变量这一改进方法应用于二代小波变换中,实现了对地震信号更加快速有效的去噪处理,并在模拟数据试算和实际数据试算中取得了良好的去噪效果。

Application of improved threshold method on second-generation wavelet transform denoising

Seismic signal denoising can effectively improve the signal-to-noise ratio and the resolution of the signal. The second-generation wavelet transform can denoise signal in different scales and resolutions and realize local analysis in which the window width can be adjusted adaptively. However, in the process of denoising, wavelet threshold method exhibits pseudo gibbs effect for discontinuous signal. This problem can be solved by the translation invariant threshold denoising method via translation-denoising-average. Therefore, based on the existing results on the second-generation wavelet transform and threshold denoising, translation invariant will be applied to the second-generation wavelet transform to denoise signal fast and effectively, and achieve good denoising result in the simulated data trial and real data trial.