Airborne Electromagnetic Data Leveling Based on Structured Model

Airborne electromagnetic method has the advantages of fast and efficient, non-contact, multi-parameter acquisition, high cost performance and strong adaptability, and has been widely used in geological mapping, mineral resources, underground water sources, environment and engineering exploration. However, because the flight mode is inevitably affected, the data difference between routes will be generated. This difference in data is represented in the imaging system as a bar graph parallel to the flight line. This structural model motivates us to build a leveling error model utilizing an anisotropic Gabor filter. Then the leveling error model is embedded into the total variational frame to avoid blind removal of leveling error. In addition, the structured variational method can also be expanded to eliminate various other types of noise with a broad prior noise assumption. We utilized the method on airborne magnetic data gathered by the Geological Survey of Ontario and juxtaposed the outcomes with published data to authenticate its reliability. Leveraging the spatial attributes of leveling error, the structured variational method proves beneficial for effectively leveling aerial geophysical data.