Study On Extraction And Identification Method Of Airborne Gamma Spectrum Anomalies In Xiangshan Volcanic Basin

Airborne gamma spectroscopy is an effective airborne geophysical prospecting method that suitable for conducting mineral surveys in large areas.It has the advantages of high detection efficiency,low measurement cost,and convenient large-area work.Among them,the extraction and identification of anomalous content data of various elements in the aerial gamma spectrum is a key link in data processing and interpretation,and it is also one of the key indicators for evaluating the pros and cons of the aerial gamma spectrum measurement results;Traditional statistics-based anomaly extraction algorithms have limitations in determining radiological anomalies,often accompanied by false anomalies;at the same time,the identification and judgment of anomalies requires manual intervention,which lacks objectivity.Fractal theory and artificial intelligence algorithms have been widely used in image and signal anomaly extraction and recognition,and have achieved good results.Thus,this paper introduces anomaly extraction technology based on fractal theory and anomaly recognition technology based on support vector machines into the airborne gamma spectrum data processing,And take the aerial gamma energy spectrum data of Xiangshan volcanic basin as an example to discuss the effectiveness and feasibility of this technology,and provide new technology and reference for the anomaly extraction and identification of aerial gamma energy spectrum data.This article first introduces the geology and radioactivity characteristics of the Xiangshan Volcanic Basin,and briefly introduces the abnormal characteristics of the airborne radioactive gamma spectrum in the study area and the processing results of traditional statistical methods.On this basis,the fractal theory and support vector machine algorithm theory are carried out.Description,focusing on its realization process and processing results.At the same time,combined with the known ore deposits and the spatial distribution of ore points in the study area,by comparing the abnormal extraction and recognition effects of various methods in the study area,it can be found that:(1)The abnormal areas extracted by traditional statistical methods are too small,and some low-value mining points in the study area cannot be identified,and useful abnormal information is easily missed.(2)Compared with traditional statistical methods,the abnormal area extracted based on fractal theory has a better corresponding relationship with the distribution of known mining points in the research area,which shows that the fractal method can effectively extract the abnormal characteristics of the aerial gamma energy spectrum,but the fractal method The extracted anomaly area is too large,although it is not easy to miss the weak anomaly area in the survey area,it also brings difficulties to the follow-up further verification work.(3)The support vector machine is more targeted for the anomaly recognition in the study area.The abnormal area identified by it corresponds well to the distribution of known mining points in the study area,and the anomaly area identified by it is also more reasonable.It proved support vector machines are reliable to identify anomalies in the study area.Compared with the conventional anomaly extraction technology,the support vector machine can more effectively identify the anomaly information in the study area.