摘要
利用小波变换研究地震勘探信号小波变换的过零点特性,本文提出了用小波变换的过零点特性和地震勘探信号相邻道的横向相关性提高信号分辨率和信噪比的新方法.该方法包括两个主要步骤:①利用相邻地震道信号具有很好相关性,而噪音相关性差的特点以及小波变换的过零点特性得到有效反射波同相轴随空间坐标的变化信息.②利用奇异值分解和最小二乘(SVD-TLS)方法沿同相轴对振幅进行多项式拟合去噪并增加信号高频提高信号分辨率.
We study the zero crossing properties of wavelet transform. A new method is developed to improve signal noise ratio and resoultion of seismic section based on the properties of wavelet transform zero-crossings and good correlations between reflected waves of neighbouring seismic traces. Firstly, the properties of wavelet transform zerocrossings are used to match two neighbouring seismic traces in order to obtain the varied information of reflector. Then, according to the matching results, the coefficients of wavelet transforms are picked out to fit a polynomial by using SVD-TLS. Thirdly, to suppress the noise the polynomial fitting values are used to replace the coefficients of wavelet transforms. finally, the energy of high frequency components of seismic trace is strengthened, and the high resolution seismic traces are reconstructed from their filtered wavelet transform coefficients by use of the inverse wavelet transforms. Processing results show that this method can improve both signal noise ratio and resolution of seismic section.
出处
《地球物理学报》
SCIE
EI
CAS
CSCD
北大核心
1997年第4期561-569,共9页
Chinese Journal of Geophysics
基金
国家自然科学基金
关键词
小波变换
过零点特性
信噪比
分辨率
地震勘探
Zero crossing property of wevelet transform, Singularity value decomposition, Total least squares, Signal noise ratio, Resolution