Zhongxun Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhongxun Wang has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Pengfei He
School of Physics and Electronic Information, Yantai University, Yantai 264005, China
2 shared publications - Jie Song
Research Institute of Information Fusion, Naval Aviation University, Yantai 265700, China
1 shared publication - Mingze Sun
Qingdao Huanghai University, Qingdao 266427, China
1 shared publication - Peixue Liu
Qingdao Huanghai University, Qingdao 266427, China
1 shared publication - Shu Liu
School of Physics and Electronic Information, Yantai University, Yantai 264005, China
1 shared publication - Tiantian Tang
Institute of Information Fusion, Naval Aviation University, Yantai 264001, China
1 shared publication
Latest Publications
- Open Access
- 9 October 2025
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 13 (10) , 1936
YOLO-PFA: Advanced Multi-Scale Feature Fusion and Dynamic Alignment for SAR Ship Detection
Maritime ship detection faces challenges due to complex object poses, variable target scales, and background interference. This paper introduces YOLO-PFA, a novel SAR ship detection model that integrates multi-scale feature fusion and dynamic alignment. By leveraging the Bidirectional Feature Pyramid Network (BiFPN), YOLO-PFA enhances cross-scale weighted feature fusion, improving detection of objects of varying sizes. The C2f-Partial Feature Aggregation (C2f-PFA) module aggregates raw and processed features, enhancing feature extraction efficiency. Furthermore, the Dynamic Alignment Detection Head (DADH) optimizes classification and regression feature interaction, enabling dynamic collaboration. Experimental results on the iVision-MRSSD dataset demonstrate YOLO-PFA’s superiority, achieving an mAP@0.5 of 95%, outperforming YOLOv11 by 1.2% and YOLOv12 by 2.8%. This paper contributes significantly to automated maritime target detection.
- 9 October 2025
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 13 (10) , 1936
Open AccessYOLO-PFA: Advanced Multi-Scale Feature Fusion and Dynamic Alignment for SAR Ship Detection
Maritime ship detection faces challenges due to complex object poses, variable target scales, and background interference. This paper introduces YOLO-PFA, a novel SAR ship detection model that integrates multi-scale feature fusion and dynamic alignment. By leveraging the Bidirectional Feature Pyramid Network (BiFPN), YOLO-PFA enhances cross-scale weighted feature fusion, improving detection of objects of varying sizes. The C2f-Partial Feature Aggregation (C2f-PFA) module aggregates raw and processed features, enhancing feature extraction efficiency. Furthermore, the Dynamic Alignment Detection Head (DADH) optimizes classification and regression feature interaction, enabling dynamic collaboration. Experimental results on the iVision-MRSSD dataset demonstrate YOLO-PFA’s superiority, achieving an mAP@0.5 of 95%, outperforming YOLOv11 by 1.2% and YOLOv12 by 2.8%. This paper contributes significantly to automated maritime target detection.
- Open Access
Receiver Location Optimization for Heterogeneous S-Band Marine Transmitters in Passive Multistatic Radar Networks via NSGA-II
Comprehensive maritime domain awareness is crucial for navigation safety, traffic management, and security surveillance. In the context of an increasingly complex modern electromagnetic environment, the disadvantages of traditional active single-station radars, such as their high cost and susceptibility to interference, have started to surface. Due to their unique advantages, such as low cost, environmental sustainability (by reusing existing signals), and resilience in congested spectral environments, non-cooperative passive multistatic radar (PMR) systems have gained significant interest in maritime monitoring. This paper presents the research background of non-cooperative passive multistatic radar systems, performs a fundamental analysis of the detection performance of multistatic radar systems, and suggests an optimization method for the transceiver configuration of non-cooperative passive multistatic radar systems based on geometric coverage theory and a signal-to-noise ratio model. A multi-objective optimization model is developed, considering both detection coverage and positioning error, and is solved using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The optimization aims to find the optimal receiver location relative to a fixed configuration of four transmitters, representing common maritime traffic patterns. According to the simulation results, the multi-target genetic algorithm can be utilized to optimize the receiver position under the S-band radar settings used in this work. Compared to a random placement baseline, this can reduce the positioning error by about 8.9% and extend the detection range by about 15.8%. Furthermore, for the specific four-transmitter configuration and S-band radar parameters considered in this study, it is found that the best detection performance is more likely to be obtained when the receiver is placed within 15 km of the transmitters’ geometric center.
Open AccessReceiver Location Optimization for Heterogeneous S-Band Marine Transmitters in Passive Multistatic Radar Networks via NSGA-II
Comprehensive maritime domain awareness is crucial for navigation safety, traffic management, and security surveillance. In the context of an increasingly complex modern electromagnetic environment, the disadvantages of traditional active single-station radars, such as their high cost and susceptibility to interference, have started to surface. Due to their unique advantages, such as low cost, environmental sustainability (by reusing existing signals), and resilience in congested spectral environments, non-cooperative passive multistatic radar (PMR) systems have gained significant interest in maritime monitoring. This paper presents the research background of non-cooperative passive multistatic radar systems, performs a fundamental analysis of the detection performance of multistatic radar systems, and suggests an optimization method for the transceiver configuration of non-cooperative passive multistatic radar systems based on geometric coverage theory and a signal-to-noise ratio model. A multi-objective optimization model is developed, considering both detection coverage and positioning error, and is solved using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The optimization aims to find the optimal receiver location relative to a fixed configuration of four transmitters, representing common maritime traffic patterns. According to the simulation results, the multi-target genetic algorithm can be utilized to optimize the receiver position under the S-band radar settings used in this work. Compared to a random placement baseline, this can reduce the positioning error by about 8.9% and extend the detection range by about 15.8%. Furthermore, for the specific four-transmitter configuration and S-band radar parameters considered in this study, it is found that the best detection performance is more likely to be obtained when the receiver is placed within 15 km of the transmitters’ geometric center.
- Open Access
- 21 September 2018
- Published by Springer Nature in EURASIP Journal on Image and Video Processing
- Vol. 2018 (1) , 92
Digital image information hiding algorithm research based on LDPC code
- 21 September 2018
- Published by Springer Nature in EURASIP Journal on Image and Video Processing
- Vol. 2018 (1) , 92
Open AccessDigital image information hiding algorithm research based on LDPC code
- Open Access
- 1 March 2017
- Published by Universitas Ahmad Dahlan in TELKOMNIKA (Telecommunication Computing Electronics and Control)
- Vol. 15 (1) , 457-463
Design of LDPC Decoder Based On FPGA in Digital Image Watermarking Technology
In this paper, a digital image watermarking system compiled by the LDPC code is introduced to embed and extract the image watermark. The system software is realized by the mixed programming with Matlab and Qt at first. It could get real-time input data for the LDPC decoder. Then the hardware of the LDPC decoder is primarily implemented by FPGA in the digital image watermarking system. And the serial port is used to make the output data of the decoder back to computer for verification. Through the simulation results, the Models time simulation diagram is given, and the watermark image compared with the original image is got. The results show that the resource usage of our system is few, and the decoding rate is fast. It has a certain practical value.
- 1 March 2017
- Published by Universitas Ahmad Dahlan in TELKOMNIKA (Telecommunication Computing Electronics and Control)
- Vol. 15 (1) , 457-463
Open AccessDesign of LDPC Decoder Based On FPGA in Digital Image Watermarking Technology
In this paper, a digital image watermarking system compiled by the LDPC code is introduced to embed and extract the image watermark. The system software is realized by the mixed programming with Matlab and Qt at first. It could get real-time input data for the LDPC decoder. Then the hardware of the LDPC decoder is primarily implemented by FPGA in the digital image watermarking system. And the serial port is used to make the output data of the decoder back to computer for verification. Through the simulation results, the Models time simulation diagram is given, and the watermark image compared with the original image is got. The results show that the resource usage of our system is few, and the decoding rate is fast. It has a certain practical value.