2D Blind-wavelet Algorithm And Its Application In The Processing Of De-noising On Metal Seismic Data

With the constant of Progress and development in the human society, peopledemand for mineral resources especially the metal mineral resources more and more.It is the main trend of exploration the mineral resources that human beings obtain themineral resources from the deep earth as the result of the mineral resources less andless on the surface of the earth and in the shallow of the earth. The geologicformations with metal deposit is very complex, and the depth of the metal ore bodywith Small size, Small scale and scattered distribution, and the stratigraphic interfaceis bad continuity, it is difficult to form the existing seismic specular reflectionconditions; The internal metal with the non-uniform liquid, weak effective signal, lowsignal noise ratio in the metal seismic data; Surface layer are filled with variablelithology and complicated structure. The traditional methods of Explorationtechniques can not meet the requirements of exploration in the depth of metallic oredeposits. Electromagnetic method in deep resolution is low, and direct current electricmethod in metal seismic exploration is shallow, so a new seismic exploration must befound that it makes the distance detection, high accuracy, high resolution.To gain the high signal to noise ratio with the data of metal seismic can bebeneficial to seismic data interpretation in the seismic data processing. However thereare serious impact on the seismic data such as Sound waves, the linear noise etc,and compared with seismic data of oil and gas, metal seismic data under the influenceof various interference factors.So it is hardly deal with all of the seismic noise withthe traditional methods of the noise reduction processing techniques in the metalseismic data. Because of lots of methods with noise processing have their ownconditions, so a new method on noise processing with metal seismic must be found,then all of the noise with metal seismic data can be eliminate.The article is analysis the characteristics of the noise with he metal seismic data, and research the method with the metal seismic data de-noising. and the work is asfollows: Research the characteristics of the distribution of the noise with the metalseismic data; Studied the related theories of the wavelet analysis on the noisereduction. Lucubrate the mallat algorithm with the two-dimensional multichannelsignals, and obtained the advantages and shortcomings of threshold de-noising ofwavelet change; Studied the related theories of the blind source separation, andobtained the advantages and shortcomings of the noise elimination with blind sourceseparation; Studied the noise processing in the metal seismic data with aone-dimensional wavelet algorithm and know its advantages and shortcomings; Though the advantages and disadvantages on the noise processing in the metalseismic data from the wavelet transform algorithm, the blind source separationalgorithm and the one-dimensional wavelet algorithm, we can obtain new algorithmnamed2d Blind-wavelet algorithm. For2d blind parameter selection and algorithm ofwavelet algorithm feasibility is discussed. Give the parameter selection and thefeasibility with the algorithm with the2d Blind-wavelet algorithm. The algorithm isperfect to remedy the insufficient about wavelet transform algorithm, blind sourceseparation algorithm, a one-dimensional wavelet algorithm, and it give full play to theadvantage of noise reduction. Metal seismic sections are clear fluid and thecharacteristic of hyperbolic is obvious after the noise reduction, it is high signal tonoise ratio and high resolution. It plays an important in exploration on deep seismicdeposit metal and has the theoretical significance and practical significance.