基于自适应K-SVD的能量泄漏恢复研究

Research on energy leakage recovery of adaptive K-SVD

针对现有常规去噪方法手段在参数选择及条件假设的不合理因素,导致去除的噪声剖面中残留部分有效信号(又称能量泄漏)的问题,本文提出了基于自适应K-SVD的能量泄漏恢复方法,实现残存有效信号的回收利用。首先采用小波阈值、f-x域去噪等传统方法处理,获得一个较高品质的样本进行字典的构造,再引入相关系数匹配准则对残差剖面进行稀疏分解,实现随机噪声与残留有效信号的二次分离。试验结果表明,本文算法能较好地兼顾地震信号的去噪与保幅,具有自适应性。

In view of the problem of residual signal(also known as energy leakage) in the noise profile of the existing conventional denoising method in the parameter selection,this paper proposes the energy leakage recovery method of adaptive K-SVD to achieve effective recycling of signals.Firstly,a high-quality sample is obtained by using traditional methods such as wavelet threshold and fx domain denoising to perform dictionary construction,and then the correlation coefficient matching criterion is introduced to sparsely decompose the residual profile so as to realize the secondary separation of random noise and residual effective signal.The experimental results show that the proposed algorithm can better balance the denoising and amplitude preservation of seismic signals and has adaptability.