Research On Edge Detection And Location Inversion Methods Of Gravity&Magnetic And Their Gradient Tensor Data

The existing automatic processing and interpretation methods for abnormal data have certain defects.For deep abnormal objects,traditional edge detection methods are prone to problems such as blur,divergence,and deformation,while the improved methods mostly use high-order derivatives,which are complex and easy to calculate.Noise interference;for positive and negative gravity anomaly data,there are problems such as false boundaries in the recognition results.When the Euler deconvolution method performs position inversion,the screening method for a large number of divergence and false inversion results relies too much on manual intervention,and the reliability is low.Therefore,the main contents of this paper are as follows:Introduce the theoretical basis of gravity,magnetic and gradient tensor data,derive the calculation formulas of the total magnetic anomaly,the three-component magnetic anomaly and the magnetic gradient tensor data of different models under the background of the magnetic field and perform forward simulation.Finally,perform the simulation results on each anomaly data The detailed characteristic analysis provides a theoretical basis for the subsequent explanatory research of this article.Organize and summarize the edge detection methods,analyze the complexity and recognition efficiency of the recognition methods;establish a deep abnormal body model,use the existing edge detection methods,and perform model simulation and comparative analysis of results under different magnetization and different noise intensity environments.Comprehensive use of image processing technology,step-by-step enhanced edge detection method to improve the boundary recognition effect,increase the resolution of target recognition,and solve the problem of divergence and deformation of the boundary in the traditional method.Using the arctangent form of the fast balanced edge detection method,and giving the sign function and high-order derivative form of the recognition formula through corresponding improvements,the balanced boundary recognition of deep and shallow abnormal targets can be realized.The method is simple and effective,and avoids The calculation of high-order derivatives is used;deep anomaly model tests and different types of measured anomaly data are used to verify the accuracy,convenience and practicability of the fast equilibrium boundary method by comparing other edge detection methods.For gravity anomaly data containing positive and negative density bodies,the improved fast equilibrium boundary identification method and high-order derivative form can avoid false boundaries caused by positive and negative anomalies.Comprehensively use the boundary recognition results and image processing methods to automatically divide the target area,and perform Euler deconvolution calculations in this area separately to reduce abnormal superimposition interference and improve the reliability of the inversion results;Quantitative screening,so as to achieve screening of inversion results,can adapt to abnormal data of different goals or scales,reduce human intervention and simplify the screening process,and ensure the quality and quantity of screening results.Through various model simulations and measured gravity anomaly data,the validity and reliability of the method are verified.