An Energy Detection Algorithm with Clustering-Based False Alarm Suppression for Magnetic Anomaly Detection.

Area of Science:

• Geophysics
• Signal Processing
• Data Science

Background:

• Orthonormal Basis Function (OBF) methods suffer from high false alarm rates and ambiguous localization in magnetic anomaly detection.
• Background noise significantly impacts the reliability of traditional magnetic anomaly detection techniques.

Purpose of the Study:

• To develop a high-confidence magnetic anomaly detection algorithm overcoming limitations of existing methods.
• To enhance target localization accuracy and reduce false alarms in noisy environments.

Main Methods:

• A novel algorithm based on hierarchical clustering with a theoretically derived optimal cut height.
• Utilizing a physical model of the magnetic dipole's vertical gradient field to determine the optimal cut height.
• Processing alarm point clouds from a Greatest-of Cell-Averaging Constant False Alarm Rate (GOCA-CFAR) detector.

Main Results:

• The algorithm effectively suppresses isolated false alarms caused by background fluctuations.
• Spatially coherent alarm clusters within the target's effective detection range are preserved.
• Demonstrated superior capability in discriminating genuine targets from false alarms compared to 1D-CFAR detection.

Conclusions:

• The proposed algorithm significantly enhances detection confidence in magnetic anomaly detection.
• Validated efficacy through simulations and field experiments.
• Provides a reliable method for distinguishing real targets from false alarms in complex environments.