A5058203513- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Vehicle License Plate Recognition
- Autonomous Vehicle Technology and Safety
- Internet Traffic Analysis and Secure E-voting
- Power Systems and Technologies
- Software-Defined Networks and 5G
- Educational Technology and Pedagogy
- Advanced Computational Techniques and Applications
- Image and Signal Denoising Methods
- Technology and Security Systems
- Healthcare Systems and Public Health
- Educational Technology and Assessment
- Network Security and Intrusion Detection
Shandong Jiaotong University
2021-2024
People's Hospital of Cangzhou
2021
Huzhou Vocational and Technical College
2021
Shaanxi Railway Institute
2012
Vehicle re-identification (Re-ID) is a challenging task that aims to recognize the same vehicle across different non-overlapping cameras. Existing attention mechanism-based methods for Re-ID often suffer from significant intra-class variation and inter-class due various factors such as illumination, occlusion, viewpoint, etc. In this paper, we propose novel network architecture Re-ID, named Dimensional Decoupling Strategy Non-local Relationship Network (DMNR-Net), which uses three modules...
Vehicle Re-identification (Re-ID) refers to finding the same vehicle shot by other cameras from a given image library, which can also be regarded as sub-problem of retrieval. It plays an important role in intelligent transportation and smart cities. The key Re-ID is extract discriminative features. To better such features improve recognition accuracy, we propose three-branch adaptive attention network—Global Relational Attention Multi-granularity Feature Learning (GRMF) feature...
OBJECTIVE To establish a nomogram model for predicting the risk of coronary artery disease in elderly patients with acute myocardial infarction (AMI). METHODS The clinical data AMI who underwent angiography department cardiology Cangzhou Central Hospital from July 2015 to March 2020 were analyzed, including age, gender, smoking history, underlying diseases, family blood pressure, left ventricular ejection fraction (LVEF), and several biochemical indicators at admission, such as total...
Ensuring the network security, resists malicious traffic attacks as much possible, and ensuring Gated Recurrent Unit (GRU) Convolutional Neural Network (CNN) are combined. Then, a Software Defined Networking (SDN) anomaly detection architecture is built continuously optimized to ensure security possible enhance reliability of architecture. The results show that proposed can greatly improve accuracy detection, its performance will be different due number CNN layers. When two-layer structure...
At present, our country is actively promoting the informatization of education, using modern information technology to change traditional education model, thereby improving quality teaching. With development open English distance project has also opened a lifelong learning system, which shows that resources system have been fully utilized. The whole people improved, and cost reduced. This great benefits people. Modern way English. main courses used in are course software required by training...
An algorithm that can effectively distinguish different vehicles with high similarity by identifying vehicle photos collected from multiple angles is called recognition algorithm. This has been extensively applicable to the region of intelligent transportation and urban computing, but it always challenging implement In this paper, an modified feature extraction method provided, which enhances traditional nonlocal neural networks improves its ability capture relationship between positions in...