Prof. Liming Fan

In order to improve the detection performance of magnetic anomaly signal with low signal-to-noise ratio (SNR), we develop an effective method using one-dimensional convolutional neural network (1D CNN) model with multi-feature fusion. In the method, the magnetic signal is processed by Hilbert-Huang transform and discrete wavelet transform to obtain its information as pre-feature in different dimensions. The 1D CNN model with three processing blocks is used to further extract features from pre-features and identified whether the anomaly signal exists or not based on multi-feature fusion. To train the model, the positive sample set is generated by simulated signals and the measured magnetic noise, while the negative sample set is only the measured magnetic noise. Simulation results show that the proposed method has high accuracies in training and test set. A field experiment is conducted to examine the detection performance of proposed method using real data. Results show that the proposed method has good detection performances in low SNR.

1 January 2022 in IEEE Geoscience and Remote Sensing Letters

Journal Article

Adaptive Magnetic Anomaly Detection Method Using Support Vector Machine

Hao Hu

Xing Liu

15 September 2020 in IEEE Sensors Journal

Journal Article

Measurement of Far Field Magnetic Moment Vector of a Moving Ferromagnetic Object

Xiaojun Zhang

Xiyuan Kang

Xin Chen

Hua Lv

Zhuoshan Geng

Chong Kang

In magnetic anomaly detection, it is a prerequisite to master the far-field magnetic moment vector information of the target. It is important to measure the magnetic moment of the target in target location, recognition and degaussing. In the paper, we proposed a method of measuring the magnetic moment vector of the target based on the total geomagnetic field. In this method, a sensor array with the scalar magnetometers was constructed and the quantitative factors of the magnetometer were analyzed. In order to eliminate the time-varying and uneven spatial distribution of geomagnetic total field, a double gradient algorithm was applied to magnetic moment measurement in this method. And a criterion function of determining the magnitude and direction of the target magnetic moment was defined. The proposed method was validated by the theoretical calculation and real experimental data. The results showed that the measured magnetic field data were qualitatively in good agreement with the theoretical curve. For the magnitude of magnetic moment, the average relative deviation was 5.1%. And the average relative deviation of inclination and declination of the moment were 3.2% and 2.5%, respectively. In addition, we attempted to separate the induced magnetic moment and the intrinsic magnetic moment of the target. The proposed method can be used as a reference for the target magnetic location and demagnetization.

4 May 2021 in IEEE Communications Letters

Journal Article

Exact Frequency Estimation in the i.i.d. Noise via KL Divergence of Accumulated Power

Shaowei Rong

To solve the frequency estimation deterioration caused by independent and identically distributed (i.i.d.) noise, this letter proposes an exact and robust frequency estimation algorithm based on the accumulated phase-difference power on Kullback-Leibler divergence (APP-KLD). First, the frequency estimation problem is modeled by the accumulated power of phase-difference between the receiving signal and Fourier basis function. Furthermore, the true frequency is estimated iteratively by maximizing the Kullback–Leibler divergence, which measures the distance between the accumulated power and a constant reference vector. Simulations demonstrate the robust performance of the APP-KLD algorithm at various levels of heavy-tailed noise and sample numbers.

15 July 2025 in Remote Sensing

Journal Article

SonarNet: Global Feature-Based Hybrid Attention Network for Side-Scan Sonar Image Segmentation

With the rapid advancement of deep learning techniques, side-scan sonar image segmentation has become a crucial task in underwater scene understanding. However, the complex and variable underwater environment poses significant challenges for salient object detection, with traditional deep learning approaches often suffering from inadequate feature representation and the loss of global context during downsampling, thus compromising the segmentation accuracy of fine structures. To address these issues, we propose SonarNet, a Global Feature-Based Hybrid Attention Network specifically designed for side-scan sonar image segmentation. SonarNet features a dual-encoder architecture that leverages residual blocks and a self-attention mechanism to simultaneously capture both global structural and local contextual information. In addition, an adaptive hybrid attention module is introduced to effectively integrate channel and spatial features, while a global enhancement block fuses multi-scale global and spatial representations from the dual encoders, mitigating information loss throughout the network. Comprehensive experiments on a dedicated underwater sonar dataset demonstrate that SonarNet outperforms ten state-of-the-art saliency detection methods, achieving a mean absolute error as low as 2.35%. These results highlight the superior performance of SonarNet in challenging sonar image segmentation tasks.

11 July 2025 in IEEE Sensors Journal

Journal Article

Magnetic Gradient for Target Localization at Point of Close Proximity Approach

Liming Fan

Zhitao Du

Menghui

Qin

Nan Li

Menghui Qin

Magnetic anomaly generated by magnetic target is widely used in many areas. The target can be located by using the magnetic anomaly. In order to decrease the complex of nonlinear function solution when locating target, we propose an effective method for target location using the close proximity approach (CPA) point on the trajectory, which is the closest point from the target to the trajectory. According to the properties of magnetic gradient at the CPA point, we derive the expressions for the target position and its magnetic moment at the CPA point. The position and magnetic moment at each measurement point be easily calculated in real-time using the expressions. Meanwhile, we also introduce an error index that can be used to determine the CPA point which is a key factor for obtaining the position and moment of target along the trajectory. Based on the CPA point determined by the index, the position and magnetic moment of target can be easily obtained. The proposed method is validated through simulation and experiment. The results demonstrate that the CPA point can be easily determined and the distance error between target and sensors array is less 10% at CPA point. Localization accuracy of the proposed method is relatively high. Therefore, the method has significant potential for target localization in mobile magnetic anomaly detection systems.

About Me

Dr. Liming Fan is an Associate Professor at the School of Marine Science and Technology, Northwestern Polytechnical University. He is a member of the Electromagnetic Committee of the China Instrument and Control Society, an expert reviewer for the National Research and Development Evaluation Center, and a youth editorial board member for several academic journals, including the Journal of Xi’an University of Posts and Telecommunications and Unmanned Systems Technology. Dr. Fan’s research focuses on underwater magnetic information sensing, magnetic field measurement technologies for unmanned marine platforms, and the design and optimization of sensor arrays. He has led and participated in more than twenty national and provincial-level research programs, including projects funded by the National Natural Science Foundation of China, national defense research initiatives, Ministry of Education international cooperation programs, and provincial natural science foundations. Dr. Fan has authored more than twenty SCI-indexed publications in internationally recognized journals such as IEEE Geoscience and Remote Sensing Letters, IEEE Sensors Journal, and Chinese Physics B.He is an active peer reviewer for multiple high-impact international journals, including IEEE GRSL, IEEE Sensors Journal, and the Journal of Sensors. His contributions reflect a sustained commitment to advancing research and innovation in marine sensing technologies and

Prof. Liming Fan

Prof. Liming Fan

Research Focus & Metrics

Academic Career Length

Publication Type Breakdown

Representative Publications

About Me