Airborne Gamma Ray Spectrum Anomaly Information Enhancement And Reconstruction Technology

Airborne Gamma Spectroscopy(AGS)measurement is carried out by mounting an airborne gamma spectroscopy survey system on an aircraft to measure the gamma ray spectrum of the ground-air interface,thereby obtaining the content(specific activity)of some uranium,thorium and potassium in the surface medium.Airborne gamma spectroscopy is one of the effective methods to explore hidden mineral resources.Compared with the geological bodies exposed on the surface,the anomalous intensity of the ore-caused airborne gamma spectroscopy on the surface is relatively weak and is easily submerged in the interference signals by the overburden or surface medium.The enhancement and reconstruction technology of data anomaly information measured by airborne gamma spectroscopy is the core technology and hot issue of data post-processing.The conventional extraction method for anomaly information of airborne gamma spectroscopy data follows the processing method of geochemical data,which processes the airborne gamma spectroscopy window data from a statistical point of view,and requires the data to conform to a normal distribution or a logarithmic normal distribution.However,in actual measurements,this perfect data does not exist.Iterative culling of outliers is required,resulting in a lack of integrity of the data.In addition,traditional statistical methods ignore the spatial properties of the data itself and the information of the geophysical fields it contains.In view of the above problems,based on the principle of airborne gamma spectroscopy measurement,this paper cut in from the perspective of spatial analysis and signal processing,and proposed a method for enhancing and reconstructing anomalous information of airborne gamma spectroscopy based on wavelet domain and singular value.The following works were carried out: first,starting from the principle of aerial gamma spectroscopy exploration,the characteristics of window data of airborne gamma spectroscopy were analyzed from the perspective of spatial analysis and signal processing.According to the characteristics of data,this paper proposed that the window data of airborne gamma spectroscopy measurement can use the idea and method of digital signal processing to process the airborne gamma window data to enhance and reconstruct the abnormal information;second,the wavelet transform method was introduced into the window data processing of the airborne gamma spectroscopy.From the perspective of wavelet domain,the strip-shaped false anomaly correction technology and the enhancement and reconstruction techniques of abnormal information were studied.Third,the singular value algorithm and multifractal theory were used to study the window data characteristics of the airborne gamma spectroscopy,and anomaly information enhancement and reconstruction model of the window data are designed according to its characteristics and the characteristics of the geophysical field in the energy domain.Through the above research work,the following research results have been achieved:(1)A line unit correction method based on one-dimensional wavelet transform is proposed to correct strip-shaped false anomalies.According to the coefficient of variation,the method uses line as the correction unit to design the filter: the wavelet basis function is rbio 3.7,use the mallat algorithm,and the optimal decomposition layer is 4.It can reduce the interference of human factors on the selection of wavelet basis functions.The results show that the method can correct the anomaly information of the line while correcting the strip-shaped false anomaly,and appropriately suppress the background.However,there are many weak anomalies in the data corrected by this method,and it is necessary to combine the other methods to eliminate the false weak anomalies.(2)The enhancement and anomaly reconstruction method based on two-dimensional wavelet data processing was adopted,and the wavelet basis function is optimized to determine the optimal decomposition layer.Still selected rbio3.7,decomposed the data,and filtered out the decomposed detailed signals of the 1,2,and 3 layer,reconstructed the approximate signals,and compared the results of enhancing and reconstructing the abnormal information.The results show that for the research area,the two-layer wavelet processing is best when performing two-dimensional wavelet decomposition.After the data was decomposed,filtered and reconstructed,the background of the measurement area was fully suppressed,weak anomalies were highlighted,and high anomalies were completely enhanced and reconstructed without distortion.(3)The original data is processed by the fractal-singular value method to enhance and reconstruct the abnormal information.Based on the fractal theory,the singular value-energy measurement spectrum was analyzed to segment the spectral line and find the intersection point according to the minimum principle of residual square sum,and calculate the critical singular value quantitatively to distinguish the complex geological structure.Through the experimental analysis of the airborne gamma energy spectrum production data,it is found that the original data is decomposed by singular values,and different singular value intervals represent different anomaly information: high singular value interval(>50)stands for background signal;medium singular value(13-50)represents mine-induced anomaly information;the low singular value portion(<13)represents other noise;and the strip-shaped false anomaly signal due to flight error exists in the high singular value.(4)Application in a certain survey area in Inner Mongolia.The paleo-uranium abundance in the survey area was calculated,the contour map and the ground gamma ray spectrum anomaly map was drawn,both of which showed that the survey area had uranium source mineralization conditions.Comparing the processing results with the paleo-uranium distribution map and the ground gamma ray spectrum anomaly verification map,it is found that the original data of the survey area was better suppressed by the above method,and the abnormal area was highlighted.Due to the weakening of the strip-like false anomalies generated by the time-domain batches,the positive and negative false anomalies appearing in the southern part of the survey area disappeared,and the anomaly range at the known ore points was reduced,and the anomaly shape is consistent with the actual ore points.