Kun Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Kun Wu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Bin Wang
Department of Automation, Tsinghua University, Beijing, China
2 shared publications - Li Yang
School of Electronic and Information, Southwest Minzu University, Chengdu, China
2 shared publications - Xiaoping Zheng
Tsinghua University
2 shared publications - Zhiyu Lu
School of Electronic and Information, Southwest Minzu University, Chengdu, China
2 shared publications - Yongxin Li
School of Electronic and Information, Southwest Minzu University, Chengdu, China
1 shared publication
Latest Publications
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
Magnetic Multitarget Recognition Based on Magnetic Gradient Tensor
The difficulty in magnetic multi-target recognition is that it is difficult to recognize magnetic targets with different magnetic moments and different depths at the same time. This is because the magnetic moment and depth of a magnetic target affect the strength of the generated magnetic field information, resulting in localization results that favor targets that generate stronger magnetic fields, while targets with weaker magnetic fields cannot be accurately identified. In addition, the traditional Euler equation-based multi-target localization method requires calibration of the sensor array before detection, which requires high calibration accuracy as it relies on the inversion of mathematical equations to solve for magnetic targets. To address this problem, this paper proposes an inverse tangent function imaging processing method based on the magnetic gradient tensor information (MGT) depth direction change rate. The method aims to balance the differences in magnetic field information between different magnetic targets and enhance the recognition of the horizontal position of targets with weaker magnetic fields. Based on this, a depth detection method based on MGT is proposed to acquire the depth information of magnetic targets. A grid-format data acquisition method is also proposed to obtain the MGT. The method does not require calibration of the sensor array or separation of the magnetic anomaly field from the total field. Simulation and experimental results show that the positioning error of this method for each magnetic target in the x, y, and z axes is less than 0.15 m. For a magnetic target with a depth of 1 m, the x-axis error is 0.15 m, the y-axis error is 0.05 m, and the z-axis error is 0.06 m. For a magnetic target with a depth of 0.7 m, the x-axis error is 0.08 m, the y-axis error is 0.12 m, and the z-axis error is 0.13 m.
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
Magnetic Multitarget Recognition Based on Magnetic Gradient Tensor
The difficulty in magnetic multi-target recognition is that it is difficult to recognize magnetic targets with different magnetic moments and different depths at the same time. This is because the magnetic moment and depth of a magnetic target affect the strength of the generated magnetic field information, resulting in localization results that favor targets that generate stronger magnetic fields, while targets with weaker magnetic fields cannot be accurately identified. In addition, the traditional Euler equation-based multi-target localization method requires calibration of the sensor array before detection, which requires high calibration accuracy as it relies on the inversion of mathematical equations to solve for magnetic targets. To address this problem, this paper proposes an inverse tangent function imaging processing method based on the magnetic gradient tensor information (MGT) depth direction change rate. The method aims to balance the differences in magnetic field information between different magnetic targets and enhance the recognition of the horizontal position of targets with weaker magnetic fields. Based on this, a depth detection method based on MGT is proposed to acquire the depth information of magnetic targets. A grid-format data acquisition method is also proposed to obtain the MGT. The method does not require calibration of the sensor array or separation of the magnetic anomaly field from the total field. Simulation and experimental results show that the positioning error of this method for each magnetic target in the x, y, and z axes is less than 0.15 m. For a magnetic target with a depth of 1 m, the x-axis error is 0.15 m, the y-axis error is 0.05 m, and the z-axis error is 0.06 m. For a magnetic target with a depth of 0.7 m, the x-axis error is 0.08 m, the y-axis error is 0.12 m, and the z-axis error is 0.13 m.
- Top Cited Papers
- 11 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-11
Predictive Model-Based Correction of Magnetic Sensor Array Sway Errors
Magnetic sensor arrays are typically used to detect magnetic targets. Currently, research on sensor array calibration focuses on solving the problems of inconsistent sensitivity, zero offset, and non-orthogonality in individual sensors, and misalignment errors between sensors. However, in magnetic field detection, sensor arrays are usually mounted on a platform or carried as a handheld device and are prone to random swaying during the detection process, leading to changes in the attitude and position of the magnetic sensors, which in turn generates swaying errors. In this study, the source of swaying error in sensor arrays was first analyzed theoretically, and a swaying error model of a magnetic sensor was established. Second, an swaying error calibration method was proposed in combination with the classical prediction model— Gaussian Process Regression (GPR), Backpropagation Neural Network (BP) and Support Vector Machine (SVM) in the field of artificial intelligence. The experimental results show that the prediction performance based on the BP neural network is the most outstanding. After correction, the relative error percentage of the swaying error in the magnetic field data decreased significantly from 165.50% to 9.35%, which is a significant improvement of the correction effect. In addition, we conducted model comparison experiments in different environments, and the results show that the BP model performs well in various environments, demonstrating its strong generalization ability and robustness. Finally, the distance error of the magnetic dipole position was significantly reduced after calibration, from 1.56 m, 1.12 m, and 2.50 m to 0.17 m, 0.04 m, and 0.04 m, respectively. Thus, the effectiveness of the calibration method was verified.
- 11 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-11
Top Cited PapersPredictive Model-Based Correction of Magnetic Sensor Array Sway Errors
Magnetic sensor arrays are typically used to detect magnetic targets. Currently, research on sensor array calibration focuses on solving the problems of inconsistent sensitivity, zero offset, and non-orthogonality in individual sensors, and misalignment errors between sensors. However, in magnetic field detection, sensor arrays are usually mounted on a platform or carried as a handheld device and are prone to random swaying during the detection process, leading to changes in the attitude and position of the magnetic sensors, which in turn generates swaying errors. In this study, the source of swaying error in sensor arrays was first analyzed theoretically, and a swaying error model of a magnetic sensor was established. Second, an swaying error calibration method was proposed in combination with the classical prediction model— Gaussian Process Regression (GPR), Backpropagation Neural Network (BP) and Support Vector Machine (SVM) in the field of artificial intelligence. The experimental results show that the prediction performance based on the BP neural network is the most outstanding. After correction, the relative error percentage of the swaying error in the magnetic field data decreased significantly from 165.50% to 9.35%, which is a significant improvement of the correction effect. In addition, we conducted model comparison experiments in different environments, and the results show that the BP model performs well in various environments, demonstrating its strong generalization ability and robustness. Finally, the distance error of the magnetic dipole position was significantly reduced after calibration, from 1.56 m, 1.12 m, and 2.50 m to 0.17 m, 0.04 m, and 0.04 m, respectively. Thus, the effectiveness of the calibration method was verified.