wu jun

Prof. Dr. wu jun

College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, China
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 February 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
YHYongxiang He
Yongxiang He
wu junwu jun
wu jun
YZYaojia Zheng
Yaojia Zheng
YZYuxin Zhang
Yuxin Zhang
hong xiaobohong xiaobo
hong xiaobo
The high-speed maglev train is a new type of transportation. The long stator track plays a critical role in the levitation guidance and traction system. Therefore, its condition directly affects the operation of maglev trains. It is extremely important to detect the abnormal condition of high-speed maglev tracks to ensure the stable, safe, and reliable operation of the train. In this article, an onboard image detection system is designed for high-speed maglev tracks, which can accurately obtain the image of long stator tracks under the harsh conditions of limited installation space, insufficient illumination, and rapid operation of vehicles. High-speed maglev trains are not yet in widespread use. In China, there is currently only one demonstration operating line located in Shanghai, and the length of the track test line is limited. Therefore, the number of track images that can be acquired is extremely limited. In view of the lack of defective samples of high-speed maglev tracks, this article proposes a data enhancement method based on sample generation and image fusion to augment the dataset of defective samples. To improve the quality of generated high-speed maglev track defect images, a joint attention layer (JEA) combining squeeze-and-exception (SE) block and spatial attention module (SAM) is designed and introduced into the generative adversarial network (GAN). This work provides a data basis for the study of track defect detection of high-speed maglev trains. In addition, this article detects the defects of high-speed maglev tracks via deep learning-based target detection algorithms, which can automatically detect, accurately classify and locate the defects of stator surface and cables, filling the gap in the field of high-speed maglev track defect detection.

29 April 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
YHYisheng Hao
Yisheng Hao
wu junwu jun
wu jun
YYYu Yao
Yu Yao
YGYue Guo
Yue Guo
To address the challenges of compromised detection accuracy caused by near-shore clutter in SAR (synthetic aperture radar) ship detection and the limited deployability of complex algorithms on embedded systems, this paper proposes L-YOLOX, a lightweight SAR target detection algorithm optimized for terminal devices. Firstly, we devise a new feature extraction module based on the MobileNetV3 block to reduce the parameters of traditional YOLOX while strengthening feature representation. Additionally, we incorporate a cross-channel local connection structure to construct an efficient lightweight feature extraction backbone, which is beneficial to improving the network’s ability to fuse SAR ship target information. Next, we develop a multi-scale detection block by using a feature pyramid architecture and dilated convolution to improve the network’s multi-scale detection performance. Finally, we integrate a lightweight convolutional attention mechanism into YOLOX’s Neck structure to enhance the expression of important target detail information and propose the Alpha-AIoU loss function to optimize the gradient propagation process and the network’s weight update. Ablation experimental results on the SSDD dataset show that our network achieves an AP of 90.8%, outperforming Baseline YOLOX with a 70.1% reduction in parameters and a 46.9% decrease in computational cost. Our network also demonstrates a marked enhancement in robustness, validating the effectiveness of our innovations. Some comparative experiments with other state-of-the-art algorithms on SSDD and HRSID further confirm the advantages of our network in terms of SAR image lightweight detection performance and generalization capacity.

1 January 2018 in Chinese Optics Letters
Verified
Journal Article
Closed access
CXChangcheng Xiang
Changcheng Xiang
CZChanghe Zhou
Changhe Zhou
WJWei Jia
Wei Jia
wu junwu jun
wu jun

10 September 2020 in Sensors
Verified
Journal Article
Open access
PDF file
hong xiaobohong xiaobo
hong xiaobo
wu junwu jun
wu jun
Yunzhou ZhangYunzhou Zhang
Yunzhou Zhang
YHYongxiang He
Yongxiang He
A novel absolute positioning sensor for high-speed maglev train based on eddy current effect is studied in this paper. The sensor is designed with photoelectric switch and four groups of unilateral coplanar code-reading detection coil combination. The photoelectric switch realizes the positioning of the marker plate, and the four groups of detection coils read the mileage code of the mileage sign plate to obtain the absolute mileage information of the vehicle, which effectively reduces the quality and volume of the sensor, and reduces the impact of ice and snow. At the same time, the code-reading reliability and speed adaptability index are proposed. The code-reading reliability of the sensor is analyzed and tested under the fluctuation of levitation guidance, and the positioning error under the speed range of 0–600 km/h is calculated and analyzed. The results show that the novel sensor has the advantages of simple and compact structure. It still satisfies the system’s requirements for absolute vehicle mileage information under the conditions of vehicle operating attitude fluctuations and changes in the full operating speed range.

13 May 2025 in Drones
Verified
Journal Article
Open access
PDF file
YYYu Yao
Yu Yao
wu junwu jun
wu jun
YHYisheng Hao
Yisheng Hao
ZHZhen Huang
Zhen Huang
ZYZixuan Yin
Zixuan Yin
JPJiahua Pi
Jiahua Pi
Due to traditional recovery systems lacking visual perception, it is difficult to monitor UAVs’ real-time status in communication-constrained or GPS-denied environments. This leads to insufficient ability in decision-making and parameter adjustment and increase uncertainty and risk of recovery. Visual inspection technology can make up for the limitations of GPS and communication and improve the autonomy and adaptability of the system. However, the existing RT-DETR algorithm is limited by single-path feature extraction, a simplified fusion mechanism, and high-frequency information loss, which makes it difficult to balance detection accuracy and computational efficiency. Therefore, this paper proposes a lightweight visual detection model based on transformer architecture to further optimize computational efficiency. Firstly, aiming at the performance bottleneck of existing models, the Parallel Backbone is proposed, which captures local features and global semantic information by sharing the initial feature extraction module and the double-branch structure, respectively, and uses the progressive fusion mechanism to realize the adaptive integration of multiscale features so as to balance the accuracy and lightness of target detection. Secondly, an adaptive multiscale feature pyramid network (AMFPN) is designed, which effectively integrates different scales of information through multi-level feature fusion and information transmission mechanism, alleviates the problem of information loss in small-target detection, and improves the detection accuracy in complex backgrounds. Finally, a wavelet frequency–domain-optimized reverse feature fusion mechanism (WT-FORM) is proposed. By using the wavelet transform to decompose the shallow features into multi-frequency bands and combining the weighted calculation and feature compensation strategy, the computational complexity is reduced, and the representation ability of the global context is further enhanced. The experimental results show that the improved model reduces the parameter size and computational load by 43.2% and 58% while maintaining detection accuracy comparable to the original RT-DETR in three datasets. Even in complex environments with low light, occlusion, or small targets, it can provide more accurate detection results.

1 July 2022 in Measurement and Control
Verified
Journal Article
Open access
YZYunzhou Zhang
Yunzhou Zhang
CCChuang Chao
Chuang Chao
wu junwu jun
wu jun
YZYuxin Zhang
Yuxin Zhang
hong xiaobohong xiaobo
hong xiaobo
A scheme of detecting magnetic field anomaly is proposed to find local short circuit faults in long stator cores of high speed maglev transit. And a composite magnetic anomaly signal processing method is designed based on wavelet with fractal, screening out a little anomalous data from massive data. First, suspicious data are quickly recognized by applying multiple fractal spectra, then being confirmed by wavelet modulus maxima calculation. The experiment results show that the method is immune to the fluctuation of 8–12 mm suspension gap and its hit rate can be improved by more than 66%.