- Advanced Neural Network Applications
- Stochastic processes and statistical mechanics
- Blockchain Technology Applications and Security
- Theoretical and Computational Physics
- Innovation in Digital Healthcare Systems
- Quality and Management Systems
- Video Surveillance and Tracking Methods
- Advanced Statistical Process Monitoring
- Autonomous Vehicle Technology and Safety
- Advanced Thermodynamics and Statistical Mechanics
- Hydrological Forecasting Using AI
- Carbon dioxide utilization in catalysis
- Robotics and Automated Systems
- Visual Attention and Saliency Detection
- Flow Measurement and Analysis
- Technology and Security Systems
- Quality and Safety in Healthcare
- Video Coding and Compression Technologies
- Smart Agriculture and AI
- Innovative Human-Technology Interaction
- Smart Grid and Power Systems
- Peer-to-Peer Network Technologies
- Image and Object Detection Techniques
- Advancements in Battery Materials
- Spectroscopy and Chemometric Analyses
State Key Laboratory of Automotive Simulation and Control
2023
First Automotive Works (China)
2023
Jilin University
2023
Shanghai Maritime University
2022-2023
Xijing University
2022
East China Normal University
2022
Shanghai Jiao Tong University
2021
Beijing Jiaotong University
2012-2020
University of Toronto
2009-2019
State Key Laboratory of Chemical Engineering
2019
Moving object segmentation (MOS), aiming at segmenting moving objects from video frames, is an important and challenging task in computer vision with various applications. With the development of deep learning (DL), MOS has also entered era models toward spatiotemporal feature learning. This paper aims to provide latest review recent DL-based methods proposed during past three years. Specifically, we present a more up-to-date categorization based on model characteristics, then compare...
In recent years, digital watermarking is playing an increasingly important role in the geographical information copyright protection. According to characteristics of GIS vector data, watermark embedding algorithms should at least meet following demands: no significant loss data accuracy, obvious reduction quality and change visual observation. this paper, watermarked inspected through polygon closure, topology, error analysis analysis. And above functions are achieved by coding. Through...
The driven diffusive system is a powerful tool to investigate properties of nonequilibrium state statistical physics, vehicle traffic, and biological transport systems. This paper presents two-channel asymmetric exclusion process model, in which collective dynamics with interactions particles between two lanes are considered. Computer simulation mean-field analysis carried out calculate the flow rate under periodic boundary. On whole, results from two-vertical-horizontal-cluster method good...
Model evaluation is critical in deep learning. However, the traditional model approach susceptible to issues of untrustworthiness, including insecure data and sharing, training, incorrect evaluation, centralized results that can be tampered easily. To minimize these untrustworthiness issues, this paper proposes a blockchain-based framework. The framework consists an access control layer, storage training layer. layer facilitates secure resource sharing. achieve fine-grained flexible control,...
In the era of deep learning as a service, ensuring that model services are sustainable is key challenge. To achieve sustainability, services, including but not limited to storage and inference, must maintain security while preserving system efficiency, be applicable all models. address these issues, we propose sub-network-based inference solution integrates blockchain IPFS, which includes highly distributed method, tamper-proof checking double-attribute-based permission management an...
This paper studies a periodic driven diffusive system, which separates into two equal-sized parts with different values of hopping rates. Competition the leads to various bulk-driven phase transitions, including shock and antishock. More interestingly, for symmetric scenario, one can observe antishock simultaneously in system. We have explained coexistence via effective boundary reservoir density. Theoretical analysis has been carried out characterize emerging nonequilibrium steady states,...
The present paper is a case study of quality control cost optimisation in an automobile body production system, where the multivariate Bayesian chart applied. Based on enterprise’s historical data, applicability proposed method analysed. function engine compartment front axle vertical beam welding clarified, and parameters are determined. Monte Carlo simulation used to testify improvement chart. result, it believed that decreasing rate false alarm important factor for economic
Abstract Currently, China’s agricultural irrigation consumes a huge amount of water, and traditional methods lead to low efficiency serious water resource waste. Agricultural management is also difficult achieve refined due the lack accurate monitoring use data information in various areas. To find more convenient flow measurement method, this paper proposes using data-driven pump characteristic parameters predict rate. Three big data-based for predicting rate well pumps were compared,...
In recent years, wireless video transmission attracts more and attention from both academia industry with the advancement of communications Internet technologies. However, by now, most studies focus on in static or low-speed mobile environment. To best our knowledge, performance high-speed environment (such as railway) has not been investigated yet. Considering unique features channels rail environment, this work, we developed a novel emulation platform to evaluate This is based software...
In this paper, we discuss how to design the gateway service mechanism for intelligent Internet of things. Particularly, is a key component in internet things, and it able perform as protocol converter construct on transport layer with high efficiency, low response time, reliability, energy consumption. Afterwards, system architecture software things are illustrated. To promote quality mechanism, novel deployment cost optimization algorithm given. Finally, an experiment prove effectiveness...
In this paper, a multivariate Bayesian control chart is designed for condition-based maintenance application. The system deterioration process modeled as 3-state hidden Markov process, with good, warning and failure states. then applied to monitor the by plotting posterior probability that in state. It has been shown literature an optimal tool statistical unlike traditional charts. new fault detection scheme developed based on average run length criterion. Comparison results former methods...
Abstract Pixel segmentation is one of the most commonly used deep learning methods for modern lane line detection. Although outperforms traditional methods, there are two main problems: slow speed and limited receptive field. In response to these problems, this paper proposes a lightweight detection algorithm based on learnable cluster self‐attention mechanism, which has extremely fast ability adapt real scenes. The process considered as clustering under row segmentation. data processed...
Electric vehicles (EVs) have considerable potential in promoting energy efficiency and carbon neutrality. State of health (SOH) estimations for battery systems can be effective avoiding accidents involving EVs. However, existing methods rarely been developed using real driving data. The complex working environments EVs their limited data acquisition capability increase the challenges estimating SOH. In this study, a novel SOH definition was established by analyzing extracting six indicators...
Simultaneous measurement of the dynamic motion each phase in dense multiphase flow is very important to understand physically structures flow. A visualization method based on image separation and reconstruction proposed measure multiple velocity fields from single-camera images. The extends conventional PIV/PTV simultaneously with a high accuracy present study. technique have wide practical application industrial processes.
The totally asymmetric exclusion process (TASEP) is an outstanding paradigm of self-driven particle models. A new two-channel TASEP model with short-range interactions in horizontal and vertical directions given. dynamic properties the system periodic boundary open are investigated by computer simulation theoretical analysis. analysis results based on cluster mean-field method agree very well simulations. flow rate decreases for strong attractive interaction, high-density phase region...
Carl Ransom Rogers (1983) attributed learning to the result of interaction learners' behavior, cognition and emotion, emphasized influence emotion on motivation interest. However, in traditional classroom teaching, teachers are unable perceive emotional state multiple students effectively, do not often pay attention their own emotions process. So it is difficult adjust students' a timely manner, effect guarantee. Therefore, very important for learners have good teaching. For students, It...