Research On Compression Algorithm Of Seismic Data And Parallelization Based On Wavelet Transform

Along with the development of seismic prospecting, the amounts of seismic data processing and explanation increase rapidly, and large numbers of redundant data are included in the seismic data transmission and access. Therefore, the reasonable compression processing to the seismic data is already an important problem.In order to achieve the certain compression ratio and satisfiy the reconstruction precision of seismic data, this paper decomposed the seismic data to obtain several sub-band by using wavelet transform based on the multi-resolution-analysis(MRA), then according to the different information which each sub-band represents, the different strategy of the quantification and coding processing can be made. This paper improved the existing compression code algorithm for seismic data, and analysed the experimental result of the existing and the improved algorithm.The experimental result indicated that the improvement algorithm is flexibler than the basic algorithm. In view of the important information region and the non-important information region, this paper adopted the different strategy of quantification and coding, which has enhanced the compression ratio and reconstruction precision of the seismic data.In the actual seismic data processing, the amounts of seismic data is generally huge, processing speed of the serial algorithm is difficult to satisfy the demand of the computation ability which grows day by day, therefore, the research of seismic data compression parallelization method has the reality significance. The paper realized the parallelization algorithm based on MPI, it has obviously reduced the computing time of compressing the mass seismic data.