A Magnetic Anomaly Detection Method Based on Variational Mode Extraction and Adaptively Weighted Orthogonal Basis Function under Magnetic Gradient Tensor

Abstract In order to improve the detection effect of Magnetic Anomaly Detection (MAD) methods on weak magnetic anomalies with low Signal-to-Noise Ratio (SNR), a method utilizing Variational Mode Extraction (VME) and adaptive weighted Orthogonal Basis Function (OBF) based on magnetic gradient tensor is proposed in this paper. Firstly, the effect of noise reduction preprocessing on the basis function matching process of the OBF method is analyzed, and based on these analyses, a new set of orthogonal basis functions is obtained using VME. Then, an adaptively assigning method of energy weight based on the magnetic angle of the observation position is proposed, which can solve the problem that the SNR of the total energy calculated by direct summation is low, due to the inconsistent contributions of different projection elements to the total energy in the full magnetic gradient OBF method. Finally, the improved orthogonal basis function method is combined with the weight assignment strategy. The results of simulations and experiments show that the strategy of weight assignment based on the magnetic angle is effective, even for weak magnetic anomalies with low SNR. The method has a good detection performance and can locate the anomalies accurately.