金属矿地震数据降噪研究——基于小波域盲分离JADE算法
|
摘要
探索深部金属矿地震勘查中数据降噪方法,提出小波域盲分离算法处理流程,通过对我国云南地区某金属矿低信噪比资料,进行实际降噪处理研究,结果表明:①将地震数据变换到小波域的盲分离JADE算法降噪处理后,再从小波域变换到时间域,比单纯的时间域内盲分离JADE算法降噪效果要好;②在降噪处理过程中,必须根据不同地区的地震数据多次试验参数组合,以其达到最佳降噪效果。将盲信号理论与小波变换有机结合进行深部金属矿地震勘查数据降噪,对利用地震方法大面积地寻找盲矿和深部隐伏矿有一定的实用价值和指导意义。
This paper explores the method of seismic exploration data noise reduction in deep metal,and proposes wavelet domain blind separation algorithm processing procedure.The actual noise reduction process and research of the low signal-to-noise ratio(SNR) in a metal mine of Yunnan Province shows that the effect would be better than JADE blind source separation algorithm noise reduction in pure time domain when the seismic data are transformed into wavelet domain blind source separation algorithm of JADE noise reduction processing,and then the wavelet domain is transformed into time domain;and in the de-noising processing,parameter combination must be tested many times according to different parts of the seismic data so as to achieve the best effect of de-noise.It is of practical value and instruction for prospecting blind ores and deep concealed ore deposits in large areas by seismic techniques to combine blind signal theory with wavelet transformation for deep seismic exploration data noise reduction.
This paper explores the method of seismic exploration data noise reduction in deep metal,and proposes wavelet domain blind separation algorithm processing procedure.The actual noise reduction process and research of the low signal-to-noise ratio(SNR) in a metal mine of Yunnan Province shows that the effect would be better than JADE blind source separation algorithm noise reduction in pure time domain when the seismic data are transformed into wavelet domain blind source separation algorithm of JADE noise reduction processing,and then the wavelet domain is transformed into time domain;and in the de-noising processing,parameter combination must be tested many times according to different parts of the seismic data so as to achieve the best effect of de-noise.It is of practical value and instruction for prospecting blind ores and deep concealed ore deposits in large areas by seismic techniques to combine blind signal theory with wavelet transformation for deep seismic exploration data noise reduction.
引文
[1]孙明,林君.金属矿地震散射波场的数值模拟研究[J].地质与勘探,2001,37(4):68-70.
[2]徐明才,高景华,柴铭涛,等.金属矿地震勘探[M].北京:地质出版社,2009.
[3]马建仓,牛奕龙,陈海洋.盲信号处理[M].北京:国防工业出版社,2006.
[4]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.
[5]袁连喜.线性盲源分离算法的理论与应用研究[D].哈尔滨:哈尔滨工程大学,2006.
[6]杨威.基于独立分量分析的地震信号随机噪声盲分离方法应用研究[D].成都:成都理工大学,2009.
[7]邓添予.基于固定点算法的地震信号随机噪声盲分离方法应用研究[D].成都:成都理工大学,2010.
[8]易叶青.基于进化算法的盲源信号分离[D].长沙:湖南大学,2005.
[9]林兰.语音信号盲分离技术的研究与应用[D].大连:大连理工大学,2006.
[10]张安清.盲分离技术及其在水声信号中的应用研究[D].大连:大连理工大学,2006.
[11]张安清,邱天爽,章新华.扩展联合对角化法的水声信号盲分离技术[J].系统工程与电子技术,2003,25(9):1 058-1 060.
[12]郭科,陈聆,陈辉,等.基于JADE算法的地震资料随机噪声盲分离方法研究[J].地学前缘,2011,18(3):302-309.
[13]倪林.小波变换与图像处理[M].合肥:中国科学技术大学出版社,2010.
[14]YANG H H,AMARI S.Adaptive on-line learningalgorithms for blind separation:Maximum entropy andminimum mutual information[J].Neural Networks.1997,9(67):1 457-1 481.
[2]徐明才,高景华,柴铭涛,等.金属矿地震勘探[M].北京:地质出版社,2009.
[3]马建仓,牛奕龙,陈海洋.盲信号处理[M].北京:国防工业出版社,2006.
[4]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.
[5]袁连喜.线性盲源分离算法的理论与应用研究[D].哈尔滨:哈尔滨工程大学,2006.
[6]杨威.基于独立分量分析的地震信号随机噪声盲分离方法应用研究[D].成都:成都理工大学,2009.
[7]邓添予.基于固定点算法的地震信号随机噪声盲分离方法应用研究[D].成都:成都理工大学,2010.
[8]易叶青.基于进化算法的盲源信号分离[D].长沙:湖南大学,2005.
[9]林兰.语音信号盲分离技术的研究与应用[D].大连:大连理工大学,2006.
[10]张安清.盲分离技术及其在水声信号中的应用研究[D].大连:大连理工大学,2006.
[11]张安清,邱天爽,章新华.扩展联合对角化法的水声信号盲分离技术[J].系统工程与电子技术,2003,25(9):1 058-1 060.
[12]郭科,陈聆,陈辉,等.基于JADE算法的地震资料随机噪声盲分离方法研究[J].地学前缘,2011,18(3):302-309.
[13]倪林.小波变换与图像处理[M].合肥:中国科学技术大学出版社,2010.
[14]YANG H H,AMARI S.Adaptive on-line learningalgorithms for blind separation:Maximum entropy andminimum mutual information[J].Neural Networks.1997,9(67):1 457-1 481.
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