Research On Compression Method Of Seismic Exploration Data

With the rapid development of seismic exploration technology,the quality of related software and hardware equipment is getting higher and higher,and the increasing volume of seismic exploration data requires higher storage and transmission requirements.Therefore,according to the characteristics of seismic exploration data,it is urgent to design an efficient compression algorithm to compress it to reduce the huge amount of data.Therefore,this paper mainly improves the lossy compression scheme of seismic exploration data based on wavelet transform and EZW algorithm,and studies and designs a new lossless compression scheme of seismic exploration data.Xu Fengtao proposed a seismic data compression algorithm based on the combination of zero-tree coding and arithmetic coding based on wavelet transform.Compared with other methods,the seismic data compression ratio is higher and the data decompression effect is better.However,Xu Fengtao only conducted research on algorithms and lacked practicality.Therefore,this article improves the compression scheme as follows and pushes its algorithm research further to practical applications: 1.Read the standard SEG-Y file and divide it into volumes The header,track header data and seismic sampling data set are two parts,the original information is not compressed for the volume header and track header data;2.The seismic sampling data set is compressed in blocks,and the data is expanded and filled according to the number of sampling points first Satisfy the conditions of multiple wavelet transforms,the value of the expanded sampling points is used as the width value of the divided block,so that the number of channels and the number of sampling points of each block meet the multiple wavelet transform conditions;It is better to expand the filling method.The experimental results show that the use of "0" value filling has a relatively minimal effect on the original Xu Fengtao's algorithm compression effect,and can still maintain a good compression effect.The compression scheme in this paper retains the good compression effect of Xu Fengtao's algorithm,while its practical application is greatly enhanced,which is very conducive to the effective use of subsequent seismic dataResearch and design a new lossless compression method for seismic exploration data,and propose a scheme for LZMA algorithm compression based on predictive coding and byte grouping.Experimental results show that the lossless compression method has advantages in compression compared to the general compression software Win RAR and Win Zip Obviously,the data volume of the original data can be further reduced by 10% to 20%,the compression performance is better,and the scheme is reasonable and feasible.In the study,it was found that the entropy value of the fourth byte of the seismic sample data set(the last 8 bits of the mantissa of IEEE 32-bit floating point data)will decrease as the dynamic range of the data decreases.In view of the poor compression effect of the pre-stack seismic sample data set in the above lossless compression scheme,a quasi-lossless compression scheme is proposed.According to the rules found in the above study,the fractional part of the seismic sample data is removed to reduce the accuracy and thus reduce its dynamic range.The rounding method is used for the fractional part,so that the reconstruction error of each seismic sample data is less than 0.5,and then the byte grouping is compressed by the LZMA algorithm;the experimental results show that the overall signal-to-noise ratio reaches 128 d B high fidelity,and the highest compression multiple About 1.5 times.Compared with other quasi-lossless compression methods based on transform coding,the compression effect is similar,but the design and calculation complexity of the quasilossless compression algorithm in this paper is greatly reduced,the compression performance is better,and the scheme is reasonable and feasible?...