盲小波算法在金属矿床地震资料去噪处理中的应用
|
摘要
根据小波分析和盲信号分离原理,提出了一种金属地震资料降噪的盲小波算法。首先将金属地震信号用小波分解为不同频带的信号;其次把不同频带的信号进行软阈值法处理,并进一步对不同频带信号进行盲分离,提取出与源信号相关的信号;最后通过小波重构估计源信号。通过盲小波算法与其他降噪技术对实际金属地震资料进行降噪处理,结果表明盲小波算法能有效消除各种干扰噪声。去噪后的金属地震资料纹理清晰,地震资料剖面信噪比显著提高。
Based on the wavelet analysis and blind source separation,this paper proposes a blind wavelet algorithm to eliminate the metal ore deposit seismic data noises.This new method contains the following three main steps.Firstly,the metal ore deposit seismic signal is decomposed into different frequency band signals by using wavelet decomposition,secondly,it needs to use soft threshold to dispose different frequency band signals,and further to separate it by the blind source separation,meanwhile the different frequency band signals are extracted through an appropriate manner,and finally,the source signal is obtained by the reconstruction signal of wavelet transform.Through the blind wavelet algorithm and the other noise reduction technology,the processing of reducing noises for the actual metal ore deposit seismic data is done.This experimental result shows that the blind wavelet algorithm can eliminate various interference noises effectively.Then,the metal ore deposit seismic data texture is clear,and the signal-to-noise ratio of the seismic data section increases significantly.
Based on the wavelet analysis and blind source separation,this paper proposes a blind wavelet algorithm to eliminate the metal ore deposit seismic data noises.This new method contains the following three main steps.Firstly,the metal ore deposit seismic signal is decomposed into different frequency band signals by using wavelet decomposition,secondly,it needs to use soft threshold to dispose different frequency band signals,and further to separate it by the blind source separation,meanwhile the different frequency band signals are extracted through an appropriate manner,and finally,the source signal is obtained by the reconstruction signal of wavelet transform.Through the blind wavelet algorithm and the other noise reduction technology,the processing of reducing noises for the actual metal ore deposit seismic data is done.This experimental result shows that the blind wavelet algorithm can eliminate various interference noises effectively.Then,the metal ore deposit seismic data texture is clear,and the signal-to-noise ratio of the seismic data section increases significantly.
引文
[1]赵天娇,何选森,陈利.基于新阈值函数小波变换的噪声盲分离算法[J].计算机应用研究,2010,27(8):2886-2888.Zhao T J,He X S,Chen L.Noisy blind source sepa-ration algorithm based on new threshold function ofwavelet transform[J].Application Research of Com-puters,2010,27(8):2886-2888.(In Chinese)
[2]Li J,Cheng C K,Jiang T Y.Wavelet de-noising ofpartial discharge signals based on genetic adaptivethreshold estimation[J].IEEE Transactions on Die-lectrics and Electrical Insulation,2012,19(2):543-548.
[3]Liua C C,Sun T Y.Heuristic wavelet shrinkage fordenoising[J].Applied Soft Computing,2011,11:256-264.
[4]Zhang H,Blackburn T R,Phung B T,et al.A novelwavelet transform technique for on-line partial dis-charge measurements:1.WT denoising algorithm[J].IEEE Trans Dielectr Electr Insul,2007,14(4):3-14.
[5]Cardoso J F,Souloumiac A.Blind beamforming fromnon-Gaussian signals[J].Proc Inst Elect Eng-F,1993,140(6):362-370.
[6]史习智.盲信号处理:理论与实践[M].上海:上海交通大学出版社,2008:47-72.Shi X Z.Blind Signal Processing:Theory and Prac-tice[M].Shanghai:Shanghai Jiaotong UniversityPress,2008:47-72.(In Chinese)
[7]Cichocki S A.自适应盲信号与图象处理[M].北京:电子工业出版社,2005:630-664.Cichocki S A.Adaptive Blind Signal and Image Pro-cessing[M].Beijing:Electronic Industry Press,2005:630-664.(In Chinese)
[8]李加文,李从心.基于频域盲解卷的噪声信号分离[J].振动与冲击,2006,25(6):100-107.Li J W,Li C X.Noise seperation based on blind de-convolution in prequency domain[J].Journal of Vi-bration and Shock,2006,25(6):100-107.(In Chi-nese)
[9]Cichocki,Amari S I.Adaptive Signal and Image Pro-cessing:Learning Algorithms and Applications[M].New York:Wiley,2002.
[10]Cardoso J F.Souloumiac A.Blind beamforming fornon-Gaussian signals[J].IEEE Proceedings F,1993,140(6):362-370.
[11]Cardoso J F.High-order for independent componentanalysis[J].Neural Computation,1999,342(1):157-192.
[12]Zbynek Koldovsky.Efficient variant of algorithmFastICA for independent component analysis attai-ning the cramer-Rao lower bound[J].IEEE Trans-actions on Neural Networks,2006,17(5):1265-1276.
[13]张念,刘天佑,李杰.FastICA算法及其在地震信号去噪中的应用[J].计算机应用研究,2009,26(4):1432-1434.Zhang N,Liu T Y,Li J.FastICA algorithm and itsapplication in seismic signal noise elimination[J].Application Research of Computers,2009,26(4):1432-1434.(In Chinese)
[2]Li J,Cheng C K,Jiang T Y.Wavelet de-noising ofpartial discharge signals based on genetic adaptivethreshold estimation[J].IEEE Transactions on Die-lectrics and Electrical Insulation,2012,19(2):543-548.
[3]Liua C C,Sun T Y.Heuristic wavelet shrinkage fordenoising[J].Applied Soft Computing,2011,11:256-264.
[4]Zhang H,Blackburn T R,Phung B T,et al.A novelwavelet transform technique for on-line partial dis-charge measurements:1.WT denoising algorithm[J].IEEE Trans Dielectr Electr Insul,2007,14(4):3-14.
[5]Cardoso J F,Souloumiac A.Blind beamforming fromnon-Gaussian signals[J].Proc Inst Elect Eng-F,1993,140(6):362-370.
[6]史习智.盲信号处理:理论与实践[M].上海:上海交通大学出版社,2008:47-72.Shi X Z.Blind Signal Processing:Theory and Prac-tice[M].Shanghai:Shanghai Jiaotong UniversityPress,2008:47-72.(In Chinese)
[7]Cichocki S A.自适应盲信号与图象处理[M].北京:电子工业出版社,2005:630-664.Cichocki S A.Adaptive Blind Signal and Image Pro-cessing[M].Beijing:Electronic Industry Press,2005:630-664.(In Chinese)
[8]李加文,李从心.基于频域盲解卷的噪声信号分离[J].振动与冲击,2006,25(6):100-107.Li J W,Li C X.Noise seperation based on blind de-convolution in prequency domain[J].Journal of Vi-bration and Shock,2006,25(6):100-107.(In Chi-nese)
[9]Cichocki,Amari S I.Adaptive Signal and Image Pro-cessing:Learning Algorithms and Applications[M].New York:Wiley,2002.
[10]Cardoso J F.Souloumiac A.Blind beamforming fornon-Gaussian signals[J].IEEE Proceedings F,1993,140(6):362-370.
[11]Cardoso J F.High-order for independent componentanalysis[J].Neural Computation,1999,342(1):157-192.
[12]Zbynek Koldovsky.Efficient variant of algorithmFastICA for independent component analysis attai-ning the cramer-Rao lower bound[J].IEEE Trans-actions on Neural Networks,2006,17(5):1265-1276.
[13]张念,刘天佑,李杰.FastICA算法及其在地震信号去噪中的应用[J].计算机应用研究,2009,26(4):1432-1434.Zhang N,Liu T Y,Li J.FastICA algorithm and itsapplication in seismic signal noise elimination[J].Application Research of Computers,2009,26(4):1432-1434.(In Chinese)
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心