Magnetic anomaly detection (MAD) technology has been widely used in geological exploration, unexploded ordnance exploration, wreck salvage, and other fields because of its attractive advantages such as all-weather implementation, high concealment, and signal propagation in water and soil. The most classic method in MAD is orthonormal basis functions (OBFs) detection which can effectively detect the magnetic target in a comparatively low signal-to-noise ratio (SNR) environment. In this article, attracted by the excellent properties of magnetic gradient tensor (MGT) invariant, the ${{\mathrm{I}}_{1}}$ -OBF and ${{\mathrm{I}}_{2}}$ -OBF methods are developed. Considering strong dependence of the ${{\mathrm{I}}_{2}}$ -OBF method on the angle between magnetic moment vector and displacement vector, and moderate detection performance of the ${{\mathrm{I}}_{1}}$ -OBF method, an effective invariant-OBF method through adaptively integrating search energy of the ${{\mathrm{I}}_{1}}$ -OBF and ${{\mathrm{I}}_{2}}$ -OBF methods is proposed. Simulations and field experiments witnessed a great improvement in the SNR of the proposed method from the perspective of signal energy. It is proved that the proposed method can stably produce promising results even under the shaking-platform condition or the specific geometric relationship between the magnetic moment vector and the displacement vector.

As vector sensors, such as the fluxgate magnetometer, are more and more widely used in magnetic anomaly signal detection and location, researchers find that the error of sensors has become an essential obstacle in the application. To solve this problem, we established a calibration model with three kinds of error. A new optimization algorithm called Partition Beetles Antennae Search is provided to solve the model parameters. Simulation results show that the new algorithm can consider both accuracy and computational complexity. Besides, we set up a single three-axis fluxgate sensor experimental platform for sensor calibration and conducted related experiments. The results further prove the effectiveness and practicability of the method. The work of this paper suggests an effective way to calibrate three-axis magnetometers.

The calibration of a three-axis magnetometer plays a vital role in its wide practical applications. Most previous calibration methods are based on the invariance of the magnitude of the magnetic field vector. However, this kind of method degrades seriously when the training data are collected at limited-range attitudes and are much of relevance. A new calibration model of magnetometer based on the combination of magnitudes of horizontal and vertical magnetic components is proposed. The numerical simulations demonstrate the proposed method can exactly estimate the error parameters both in the case of small and large-range rotation angle. The result of the physical experiment is consistent with simulations, showing the method in this paper has a better calibration performance than previous methods when the range of attitude angle of the magnetometer is limited.

With low cost, simple structure and high accuracy, the three-axis fluxgate magnetometer (TFM) has been frequently used in many fields. However, the systematic errors are a major obstacle to the performance of TFM. Based on Taylor expansion and Cholesky decomposition, a new linear error calibration model is proposed from the general error model of TFM. In our new model, the parameter matrix can be decomposed to two parts: one is completely equivalent to the parameter matrix of the traditional linear model, and the other is a matrix containing second-order terms composed of error parameters, which was neglected in the traditional linear model. Thus, compared with the traditional model, our new linear model can acquire better calibration performances. In simulations, the functional link artificial neural network algorithm is provided to solve the model parameters. Simulation results show that the new model improves the calibration effect by 32.9 to 33.8% compared with the existing linear calibration model. After that, a single three-axis fluxgate magnetometer experimental platform is set up and related physical experiment is carried out. The results further prove the advantages and the practical value of the proposed model. Compared with the reference method, the effect is improved by 79.3%. The conclusions of this paper can expand the linear calibration method of three-axis magnetometer and improve the accuracy of such methods.

It has been difficult to achieve a suitable balance between effectiveness and efficiency in lightweight semantic segmentation networks in recent years. The goal of this work is to present an efficient and reliable semantic segmentation method called EBUNet, which is aimed at achieving a favorable trade-off between inference speed and prediction accuracy. Initially, we develop an Efficient Bottleneck Unit (EBU) that employs depth-wise convolution and depth-wise dilated convolution to obtain adequate features with moderate computation costs. Then, we developed a novel Image Partition Attention Module (IPAM), which divides feature maps into subregions and generates attention weights based on them. As a third step, we developed a novel lightweight attention decoder with which to retrieve spatial information effectively. Extensive experiments show that our EBUNet achieves 73.4% mIou and 152 FPS on the Cityscapes dataset and 72.2% mIoU and 147 FPS on the Camvid dataset with only 1.57 M parameters. The results of the experiment confirm that the proposed model is capable of making a decent trade-off in terms of accuracy, inference, and model size. The source code of our EBUNet is available at (https://github.com/Skybird1101/EBUNet).

When using the technique of magnetic gradient tensor measurements to obtain the position of magnetic objects, calibration of the magnetic tensor gradiometer plays a pivotal role in precisely locating the target, and extensive research has been carried out on this up to now. However, previous studies have always lacked sufficient discussion on the position error of magnetometers in magnetic tensor gradiometers caused by inaccurate installment of magnetometers. In this paper, we analyze and correct this position error based on a magnetic dipole source. The result of the simulation exemplifies that the magnetometer’s position error will affect the locating accuracy and, therefore, it is worth correcting this error. The relationship between position error and magnetic gradient tensor components is established, followed by an error correction method based on this relationship. Simulations illustrate that this method can effectively decrease the effect caused by the position error of magnetometers and improve the locating performance with locating error and magnetic moment errors dropping from 2 to 0.2 m and $6\times {10}^5\mathrm{A}\cdot {\mathrm{m}}^2$ to $5\times {10}^4\mathrm{A}\cdot {\mathrm{m}}^2$ , respectively.