Magnetic Anomaly Feature Extraction Using Group Sparse via Ensemble Generalized Minimax--Concave Penalty Based on the Tunable Q-factor Wavelet Transform

Extracting the accurate magnetic anomaly signal from background noise is an essential step in magnetic anomaly detection. A group-sparse feature extraction method based on the tunable Q-factor wavelet transform and generalized minimax-concave penalty is developed in this paper. Compared with the traditional wavelet transform, the proposed method significantly improves the sparsity of the extracted features. The tunable Q-factor wavelet transform removes low-frequency interference through the selection of the wavelet band. The generalized minimax-concave penalty function uses an adjustable parameter to control the degree of non-convexity in the regularization term, to realize the controllable global convex optimization enabling deal with the residual noise from the reconstructed signals with group property. The practical experiment verifies the effectiveness of the proposed method in magnetic anomaly detection.