Research On Seismic Instantaneous Attribute Extraction Method Based On Hilbert Huang Transform

Seismic data is a kind of signal which is non-stationary and nonlinear and formed by the artificial excitation of seismic waves and the earth filtering. The traditional analysis methods of signal are mainly based on the Fourier transform, which is suitable for the analysis and processing of stationary signals. When it comes to the analysis and processing of seismic signals, the Fourier transform can't get a satisfactory result. Hilbert Huang transform is a new method of analyzing non-stationary and nonlinear signal in recent years. It not only gets rid of the limitation of the Heisenberg's uncertainty principle in the linear time-frequency analysis method, but also avoids the interference of cross terms in quadratic time-frequency analysis method. Empirical mode decomposition is one of the most important steps in the Hilbert Huang transform and the signal will be self-adaptively decomposed into multiple component signals which is called intrinsic mode function by empirical mode decomposition. Then these component signals can obtain instantaneous parameters which have definite physical meaning by Hilbert transform.The research and application of seismic instantaneous attributes is an important part of seismic exploration methods. The seismic instantaneous attributes (mainly refers to the instantaneous amplitude, instantaneous phase, instantaneous frequency, namely " three-instantaneous " attributes) contains important feature information. It is extremely useful for understanding of geological features correctly and detailed seismologic interpretation. Hilbert transform and wavelet transform are two common methods which are used for analytic signal analysis. But the seismic instantaneous attributes extraction has their own shortcomings based on the two methods and Empirical mode decomposition method can better overcome the weakness of Hilbert transform. However, in practical application, there are still some problems about the empirical mode decomposition, such as end effect, termination criterion and mode mixing.Firstly, the author of this paper investigated and studied the seismic instantaneous attributes extraction method and made a comparative analysis about the defects of complex seismic trace analysis between the methods of Hilbert transform and wavelet transform, introduced into the intrinsic mode function by the concept of instantaneous frequency. Secondly he introduces the basic concept and realization method of the empirical mode decomposition, analyzed the application effect and problems of digital signal applied by the method, studied the improved algorithm of ensemble empirical mode decomposition and complementary ensemble empirical mode decomposition. By using the improved algorithm to extract the seismic instantaneous attributes and applied to the thickness of sand body analysis simply, certain results have been achieved.