Yan Xu

Prof. Dr. Yan Xu


Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
12 October 2021 in Scientific Reports
Verified
Journal Article
Open access
PDF file
LRLeilei Rong
Leilei Rong
YXYan Xu
Yan Xu
XZXiaolei Zhou
Xiaolei Zhou
LHLisu Han
Lisu Han
LLLinghui Li
Linghui Li
XPXuguang Pan
Xuguang Pan
Vehicle re-identification (re-id) aims to solve the problems of matching and identifying the same vehicle under the scenes across multiple surveillance cameras. For public security and intelligent transportation system (ITS), it is extremely important to locate the target vehicle quickly and accurately in the massive vehicle database. However, re-id of the target vehicle is very challenging due to many factors, such as the orientation variations, illumination changes, occlusion, low resolution, rapid vehicle movement, and amounts of similar vehicle models. In order to resolve the difficulties and enhance the accuracy for vehicle re-id, in this work, we propose an improved multi-branch network in which global–local feature fusion, channel attention mechanism and weighted local feature are comprehensively combined. Firstly, the fusion of global and local features is adopted to obtain more information of the vehicle and enhance the learning ability of the model; Secondly, the channel attention module in the feature extraction branch is embedded to extract the personalized features of the targeting vehicle; Finally, the background and noise information on feature extraction is controlled by weighted local feature. The results of comprehensive experiments on the mainstream evaluation datasets including VeRi-776, VRIC, and VehicleID indicate that our method can effectively improve the accuracy of vehicle re-identification and is superior to the state-of-the-art methods.

8 May 2019 in RSC Advances
Verified
Journal Article
Open access
PDF file
YYYongping Yao
Yongping Yao
XLXiaowen Li
Xiaowen Li
RSRengang Song
Rengang Song
NCNa Cui
Na Cui
SLShande Liu
Shande Liu
JHJingliang He
Jingliang He
In this paper, we prepared a high-quality multilayer ReSe 2 saturable absorber with a liquid-phase exfoliation method and characterized its saturable absorption properties around 2 μm.

28 September 2024 in Micromachines
Verified
Journal Article
Open access
PDF file
MXMaosen Xu
Maosen Xu
WTWei Tian
Wei Tian
YLYuzhe Lin
Yuzhe Lin
YXYan Xu
Yan Xu
JTJifang Tao
Jifang Tao
A carbon dioxide (CO2) gas sensor based on non-dispersive infrared (NDIR) technology has been developed and is suitable for use in portable devices for high-precision CO2 detection. The NDIR gas sensor comprises a MEMS infrared emitter, a MEMS thermopile detector with an integrated optical filter, and a compact gas cell with high optical coupling efficiency. A dual-ellipsoid mirror optical system was designed, and based on optical simulation analysis, the structure of the dual-ellipsoid reflective gas chamber was designed and optimized, achieving a coupling efficiency of up to 54%. Optical and thermal simulations were conducted to design the sensor structure, considering thermal management and light analysis. By optimizing the gas cell structure and conditioning circuit, we effectively reduced the sensor’s baseline noise, enhancing the overall reliability and stability of the system. The sensor’s dimensions were 20 mm × 10 mm × 4 mm (L × W × H), only 15% of the size of traditional NDIR gas sensors with equivalent detection resolution. The developed sensor offers high sensitivity and low noise, with a sensitivity of 15 μV/ppm, a detection limit of 90 ppm, and a resolution of 30 ppm. The total power consumption of the whole sensor system is 6.5 mW, with a maximum power consumption of only 90 mW.

10 January 2024 in High Voltage
Verified
Journal Article
Open access
XYXiangyu Yang
Xiangyu Yang
YTYouping Tu
Youping Tu
NUNorth China Electric University
North China Electric University
ZZZhong Zheng
Zhong Zheng
GCGeng Chen
Geng Chen
YXYan Xu
Yan Xu
The strain clamps and leading wires are important components that connect conductors on overhead transmission lines and conduct current. During operation, poor contact between these components can cause abnormal overheating, leading to electric failures and threatening power system reliability. Recently, the use of unmanned aerial vehicles equipped with infrared thermal imagers for strain clamp and leading wire maintenance has become increasingly popular. Deep learning‐based image recognition shows promising prospects for intelligent fault diagnosis of overheating faults. A pre‐treatment method is proposed based on dynamic histogram equalisation to enhance the contrast of infrared images. The DeepLab v3+ network, loss function, and existing networks with different backbones are compared. The DeepLab v3+ network with ResNet101 and convolutional block attention module added, and the Focal Loss function achieved the highest performance with an average pixel accuracy of 0.614, an average intersection over union ( AIoU ) of 0.567, an F1 score of 0.644, and a frequency weighted intersection over union of 0.594 on the test set. The optimised Atrous rates has increased the AIoU by 12.91%. Moreover, an intelligent diagnosis scheme for evaluating the defect state of the strain clamps and leading wires is proposed and which achieves a diagnostic accuracy of 91.0%.

23 January 2024 in Scientific Reports
Verified
Journal Article
Open access
PDF file
XGXiaoyan Guo
Xiaoyan Guo
JYJie Yang
Jie Yang
XJXinyu Jia
Xinyu Jia
CZChuanyan Zang
Chuanyan Zang
YXYan Xu
Yan Xu
ZCZhaoyang Chen
Zhaoyang Chen
Vehicle re-identification (Re-ID) involves identifying the same vehicle captured by other cameras, given a vehicle image. It plays a crucial role in the development of safe cities and smart cities. With the rapid growth and implementation of unmanned aerial vehicles (UAVs) technology, vehicle Re-ID in UAV aerial photography scenes has garnered significant attention from researchers. However, due to the high altitude of UAVs, the shooting angle of vehicle images sometimes approximates vertical, resulting in fewer local features for Re-ID. Therefore, this paper proposes a novel dual-pooling attention (DpA) module, which achieves the extraction and enhancement of locally important information about vehicles from both channel and spatial dimensions by constructing two branches of channel-pooling attention (CpA) and spatial-pooling attention (SpA), and employing multiple pooling operations to enhance the attention to fine-grained information of vehicles. Specifically, the CpA module operates between the channels of the feature map and splices features by combining four pooling operations so that vehicle regions containing discriminative information are given greater attention. The SpA module uses the same pooling operations strategy to identify discriminative representations and merge vehicle features in image regions in a weighted manner. The feature information of both dimensions is finally fused and trained jointly using label smoothing cross-entropy loss and hard mining triplet loss, thus solving the problem of missing detail information due to the high height of UAV shots. The proposed method’s effectiveness is demonstrated through extensive experiments on the UAV-based vehicle datasets VeRi-UAV and VRU.

30 March 2024 in Micromachines
Verified
Journal Article
Open access
PDF file
MXMaosen Xu
Maosen Xu
XTXin Tian
Xin Tian
YLYuzhe Lin
Yuzhe Lin
YXYan Xu
Yan Xu
JTJifang Tao
Jifang Tao
Ozone (O3) is a critical gas in various industrial applications, particularly in semiconductor manufacturing, where it is used for wafer cleaning and oxidation processes. Accurate and reliable detection of ozone concentration is essential for process control, ensuring product quality, and safeguarding workplace safety. By studying the UV absorption characteristics of O3 and combining the specific operational needs of semiconductor process gas analysis, a pressure-insensitive ozone gas sensor has been developed. In its optical structure, a straight-through design without corners was adopted, achieving a coupling efficiency of 52% in the gas chamber. This device can operate reliably in a temperature range from 0 °C to 50 °C, with only ±0.3% full-scale error across the entire temperature range. The sensor consists of a deep ultraviolet light-emitting diode in a narrow spectrum centered at 254 nm, a photodetector, and a gas chamber, with dimensions of 85 mm × 25 mm × 35 mm. The performance of the sensor has been meticulously evaluated through simulation and experimental analysis. The sensor’s gas detection accuracy is 750 ppb, with a rapid response time (t90) of 7 s, and a limit of detection of 2.26 ppm. It has the potential to be applied in various fields for ozone monitoring, including the semiconductor industry, water treatment facilities, and environmental research.