A5100456144- Cooperative Communication and Network Coding
- Topic Modeling
- Advanced Graph Neural Networks
- Advanced MIMO Systems Optimization
- Natural Language Processing Techniques
- Anomaly Detection Techniques and Applications
- Complex Network Analysis Techniques
- Advanced Wireless Network Optimization
- Advanced Wireless Communication Techniques
- Semantic Web and Ontologies
- Advanced Text Analysis Techniques
- Advanced Computational Techniques and Applications
- Cognitive Computing and Networks
- Misinformation and Its Impacts
- Laser-induced spectroscopy and plasma
- Biomedical Text Mining and Ontologies
- Image Retrieval and Classification Techniques
- Explainable Artificial Intelligence (XAI)
- Advanced Image and Video Retrieval Techniques
- Full-Duplex Wireless Communications
- Blind Source Separation Techniques
- Physical Unclonable Functions (PUFs) and Hardware Security
- VLSI and Analog Circuit Testing
- Integrated Circuits and Semiconductor Failure Analysis
- Adversarial Robustness in Machine Learning
Xidian University
2013-2025
Central South University
2025
Zigong First People's Hospital
2025
Hubei University
2007-2024
Jilin Electric Power Research Institute (China)
2024
Guangxi University
2024
Guangxi University of Chinese Medicine
2024
Anhui University
2023
Xi'an Technological University
2009-2023
Peking University
2011-2022
Distracted driving causes a large number of traffic accident fatalities and is becoming an increasingly important issue in recent research on safety. Gesture patterns are less distinguishable vehicles due to in-vehicle physical constraints body occlusions from the drivers. However, by capitalizing modern camera technology, convolutional neural network (CNN) can be used for visual analysis. In this paper, we present hybrid CNN framework (HCF) detect behaviors distracted drivers using deep...
Neural network techniques are widely used in embedding, boosting the result of node classification, link prediction, visualization and other tasks both aspects efficiency quality. All state art algorithms put effort on neighborhood information try to make full use it. However, it is hard recognize core periphery structures simply based neighborhood. In this paper, we first discuss influence brought by random-walk sampling strategies embedding results. Theoretical experimental evidences show...
The extraction of biomarkers from functional connectivity (FC) in the brain is great significance for diagnosis mental disorders. In recent years, with development deep learning, several methods have been proposed to assist depression and promote its automatic identification. However, these still some limitations. current approaches overlook importance subgraphs graphs, resulting low accuracy. Using accuracy FC analysis may lead unreliable results. To address issues, we designed a graph...
Three isostructural cyano-bridged 3d-4f compounds, [YFe(CN)6(hep)2(H2O)4] (1), [DyFe(CN)6(hep)2(H2O)4] (2), and [DyCo(CN)6(hep)2(H2O)4] (3), were successfully assembled by site-targeted substitution of the 3d or rare-earth ions. All compounds have been structurally characterized to display slightly distorted pentagonal-bipyramidal local coordination geometry around Magnetic analyses revealed negligible magnetic coupling in compound 1, antiferromagnetic intradimer interaction 2, weak...
The sensors can collect the athletes' movement data in real time, providing more accurate and objective feedback for training process. This study aims to explore application of video behavior rapid detection based on sports wearable sensor devices training. By integrating information, comprehensive behavioral analysis is provided help coaches better understand athlete performance progress. uses an approach motion technology, which athletes wear exercise during Sensors automatically record...
Named Entity Recognition for power data refers to the identification of key designative contents, such as equipment names, operation data, etc., from text domain achieve extraction and classification information perspective expertise. Deep neural networks have shown great effectiveness in system NER. However, Chinese NER suffers problems insufficient training a wide variety entities. To solve these problems, we propose entity recognition model based on Lexical enhancement Global pointer....
Previous works on knowledge-to-text generation take as input a few RDF triples or key-value pairs conveying the knowledge of some entities to generate natural language description. Existing datasets, such WIKIBIO, WebNLG, and E2E, basically have good alignment between an triple/pair set its output text. However, in practice, could be more than enough, since description may only cover most significant knowledge. In this paper, we introduce large-scale challenging dataset facilitate study...
Two novel full-length P450 cDNAs, CYP6CS1 and CYP6CW1, were cloned from the fourth instar nymphs of brown planthopper Nilaparvata lugens Stål (Hemiptera: Delphacidae) reared on its susceptible rice variety Taichung Native 1 (TN1) plants. The deduced proteins are typical microsomal P450s sharing conserved structural functional domains with other insect CYP6 members. Temporal expression analysis by northern blot hybridization indicated pre-exposure to N. moderately resistant Minghui 63 (MH63)...
This paper reports the results of SHREC'15 track: 3D Object Retrieval with Multimodal Views, which goal is to evaluate performance retrieval algorithms when multimodal views are employed for object representation. In this task, a collection 505 objects generated and both color images depth provided each object. 311 selected as queries average measured. The track attracted six participants submission 26 runs, two tasks. evaluation show promising scenario about view-based methods, reveal...
In this paper, we study the reliable packet forwarding in Wireless Sensor Networks (WSNs) with multiple-input multiple-output (MIMO) and orthogonal space time block codes (OSTBC) techniques. The objective is to propose a cross-layer optimized scheme maximize Successful Transmission Rate (STR) while satisfying given end-to-end power consumption constraint. channel coding, allocation, route planning are jointly considered significantly improve transmission quality terms of STR. joint...
We demonstrate two distinct experimental processes involving the large-area growth of ordered and disordered silicon nanowire arrays (SiNWs) on a p-type substrate using metal-assisted chemical etching method. The are based monocrystalline wafers by randomly distributed Ag films ultra-thin Au with nano-mesh arrays, respectively, wherein SiNWs is implemented specific proportion HF-containing solution at room temperature. In this study, microstructural change mechanisms for morphologically...