论文部分内容阅读
叠前地震资料是多种波的复合体,为了提取有效波,必需消除主要的干扰波,如面波、高频随机干扰波。一般的去噪方法有:一维滤波、滤波、变换、f—x预测等,但它们都侧重于考虑某一方面的单一特性或某种条件假设,这些方法具有一定的局限性。小波包变换是一种时频分析的方法,它对地震资料进行时频精细划分优于小波变换,与传统的Fourier变换相比,它能刻画出具有相同频率的有效波与干扰波在时间-空间域的分布。经过小波包对叠前资料分解,可分离出面波、高频随机干扰等,然后再经过小波包重构,可有效地剔除干扰,且对有效波的伤害较少。经实际资料验证,小波包变换,确实是一个十分有效的去噪方法。
Prestack seismic data is a complex of many waves. In order to extract effective waves, it is necessary to eliminate the main interference waves such as surface waves and high-frequency random disturbances. The general denoising methods are: one-dimensional filtering, filtering, transformation, f-x prediction, etc. But all of them focus on considering a single characteristic or certain conditional assumptions in one aspect. These methods have some limitations. Wavelet packet transform is a method of time-frequency analysis. It performs the time-frequency fine division of seismic data better than wavelet transform. Compared with the traditional Fourier transform, wavelet packet transform can describe the effective wave and the interference wave with the same frequency in the time- Spatial domain distribution. After wavelet packet pre-stack data decomposition, surface wave and high-frequency random interference can be separated and then reconstructed by wavelet packet, the interference can be effectively eliminated and the damage to the effective wave is less. The actual data validation, wavelet packet transform, is indeed a very effective denoising method.