Preprocessed Method and Application of Magnetic Gradient Tensor Data

Jinpeng Li,Yingtang Zhang,Hongbo Fan,Zhining Li

doi:10.1109/access.2019.2956057

Jinpeng Li, Yingtang Zhang + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2956057

Copy DOI

Abstract

The edge detection method is significant in the interpretation of magnetic data, and can effectively identify the edge of the magnetic source. In this paper, a new edge detection method of magnetic sources is proposed under the oblique magnetization condition. First, the data of the magnetic field component is transformed into the total magnitude anomaly and its direction cosinesaccording to the inter-relationship between different magnetic field components. To eliminate noise, we use the K-Singular Value Decomposition denoising method based on the mean signal improvement to reconstruct the magnetic signal. Then, the three components of the magnetic field data is converted into the data of three components from the same magnetic source under the vertical magnetization condition by using the calculated magnetization direction of the magnetic source. Finally, the edge of the magnetic source is calculated through the ratio of the magnetic gradient tensor invariant to the B <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">zz</sub> component data under the vertical magnetization condition. This method is verified by simulation and experiment. The simulation and experimental results obtained show that this method considerably reduces the effects of the oblique magnetization. Besides, the method in this paper has high noise immunity.

Highlights

In recent years, magnetic data interpretation based on magnetic gradient tensor plays an increasingly important role in military, environment and resource exploration
We have proposed a new edge detection method of magnetic sources under the oblique magnetization condition
The Total horizontal derivative (THD), Theta, Tilt angle, and invariant edge detection (IED) maps obtained by using noisy magnetic gradient tensor components of the rectangular prism under the oblique magnetization condition are displayed in Fig. 7 (a), (b), (c) and (d); the IED map obtained by using the preprocessed method in this paper is shown in Fig. 7 (e) and (f )

Summary

INTRODUCTION

Magnetic data interpretation based on magnetic gradient tensor plays an increasingly important role in military (unexploded ordnance detection), environment and resource exploration. We have proposed a new edge detection method of magnetic sources under the oblique magnetization condition. The preprocessed process of the magnetic gradient tensor and the edge detection method are described in the section 2. B. PREPROCESSED METHOD OF THE DATA OF THE MAGNETIC GRADIENT TENSOR 1) K-SINGULAR VALUE DECOMPOSITION DENOISING METHOD BASED ON MEAN SIGNAL IMPROVEMENT Since noise has a strong influence on the application of the data of the magnetic gradient tensor, reducing noise in the magnetic field data is needed. We propose a KSVD denoising method based on the mean signal (KSVD-MS) improvement to reconstruct the magnetic data in this paper. Components of the magnetic field data are transformed into the total magnitude anomaly and its direction cosines to preserve the inter-relationship information between different magnetic field components while removing noises. N where Bx , By, Bz are denoised components of magnetic field

APPLICATION OF MAGNETIC GRADIENT TENSOR IN MAGNETIC SOURCE EDGE DETECTION

TEST ON THE REAL MAGNETIC DATA

CONCLUSIONS