小波变换压制噪声在单道地震资料处理中的应用

Application of wavelet transform for noise suppression in single-channel seismic data processing

海洋区域地质调查中,单道地震勘探以其高分辨率、方便快捷等优点成为揭示浅地层地质构造的重要手段之一.结合前人研究成果并出于保护信号的完整性,对低频部分的噪声采用带通滤波进行滤除,而对高频信号中的随机噪声采用小波阈值去噪方法进行压制处理.在小波阈值去噪过程中,通过比较选用最佳小波基函数sym5对地震剖面进行三层小波分解.结果表明:两种阈值去噪方法在带通滤波的基础上进一步提升了剖面的质量,使得反射层面变得连续可追踪.软阈值去噪滤掉了剖面上的较多信息,去噪效果较为彻底,硬阈值去噪则保留了较多的细节信息.

Single-channel seismic exploration has become one of important means to reveal shallow geologic structure in marine geological investigations with its advantages of easy operation,high-efficient and low-cost. Based on previous research results and for the protection of the integrity of signal,the band-pass filter to filter out the random noise in the low-frequency signal and the wavelet threshold de-noising method to filter the random noise in the high-frequency signal is used. In the experiments,the sym5 wavelet is used to decompose the seismic profile by three layers of wavelet decomposition with soft and hard threshold de-noising. The results show that two threshold de-noising methods further enhance the quality of the section on the basis of the band-pass filter,which makes the reflection event become continuous tracking. The soft threshold de-noising method filters out more information on the profile,and the de-noising method effect is more thorough. And the hard threshold de-noising method retains more details.