A5101678402- Service-Oriented Architecture and Web Services
- Business Process Modeling and Analysis
- Anomaly Detection Techniques and Applications
- Advanced Image Fusion Techniques
- 3D Shape Modeling and Analysis
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Peer-to-Peer Network Technologies
- Evacuation and Crowd Dynamics
- Image Processing and 3D Reconstruction
- Mobile Agent-Based Network Management
- Collaboration in agile enterprises
- Data Mining Algorithms and Applications
- Advanced Algorithms and Applications
- Remote-Sensing Image Classification
- Data Management and Algorithms
- E-commerce and Technology Innovations
- Video Analysis and Summarization
- Traffic control and management
- Advanced Computational Techniques and Applications
- Image and Signal Denoising Methods
- Transportation Planning and Optimization
- Human Pose and Action Recognition
- Metaheuristic Optimization Algorithms Research
- Access Control and Trust
Hunan University
2025
Tsinghua University
2013-2024
China Medical University
2024
First Hospital of China Medical University
2024
Institute of High Performance Computing
2022-2024
Agency for Science, Technology and Research
2024
Shanghai University of Traditional Chinese Medicine
2024
NARI Group (China)
2020-2022
Xi'an Jiaotong University
2004-2018
Powertech Labs (Canada)
2018
In recent decades, Computer aided retrieval, analysis of text, image data and graphical technology are developing rapidly. However, the development in field 3D model retrieval is still its infant stage. Thus, related technologies, such as preprocess, normalization, extracting descriptor, index, similarity measure get a long way to explore. this paper, we review various methods content based propose taxonomy for these methods. We also evaluate several requirements shape representation: (1)...
With the prevalence of smart devices, such as phones, wearable equipments, and infrastructures, location-based service (LBS) has thrived in our daily life. In those practical LBS applications, group detection tracking is a context-related research field many scenarios, school yard, office building, shopping mall so on. this paper, we heuristically develop temporal-spatial method for clustering locating groups, then leverage CRF-based event mechanism to improve performance recognizing...
Vessel traffic services (VTS) systems are critical to safeguard maritime safety in which sensing and information technologies implemented assist gathering, storing, processing, visualizing data inclusively produced as a system solution. Mariners all over the world recognize VTS aids navigation. For instance, they use provided by request availability of pilots when vessels arriving at port or while transiting waterways closer shoreline country. However, current still work "passive" mode,...
The structure and culture of an organization does affect implementation projects. In this paper we try to identify organizational factors that could the efforts Integrated Financial Management Information System (IFMIS). information system in question has taken overtly a long time it's not complete yet. We set out And whether issues are at play particular project. project under study is large-scale integrated which aims strengthening further developing wider public service Kenya. borrow...
Image fusion rule is the key that influences quality of image which a primary branch multi-sensor information fusion. This paper proposes based on combination PCA and consistency checking applies it to multi-scale algorithm. The method approximation decomposed by adopts "select max" detail image, later checked with decision. proposed overcomes instability incontinuity occurring in before. experimental results show better than traditional rules, this can effectively improve fusion, especially...
Skyline, aiming at finding a Pareto optimal subset of points in multi-dimensional dataset, has gained great interest due to its extensive use for multi-criteria analysis and decision making. Skyline consists all that are not dominated by, or worse than other points. It is candidate set solution, which depends on specific evaluation criterion optimum. However, conventional skyline queries, return individual points, inadequate group querying case since combinations required. To address this...
In this paper, an extended FF model (floor field model) to simulate pedestrian dynamics in complex scenarios is proposed. Considering that pedestrians are unaware of the global view traffic path, we introduce pedestrians' local views and propose a framework change pedestrian's static floor each time they enter new convex area. A limited polygon. When travel from one area another, make decisions about next goal according distances between them candidate goals, as well densities capacity...
Recently, significant progress has been made in time series classification with deep learning. However, using learning models to solve generally suffers from expensive calculations and difficulty of data labeling. In this work, we study self-supervised pre-training overcome these challenges. Compared the existing works, focus on universal unlabeled pretraining. To end, propose a novel end-to-end neural network architecture based self-attention, which is suitable for capturing long-term...
Youzhi Zheng Tsinghua University, Department of Computer Science and Technology, Beijing, China 100084 Qin School Software, We propose a novel multiscale decomposition (MSD) image fusion algorithm, which is region-based using bidimensional empirical mode (BEMD). BEMD 2-D data-driven derived from the (EMD), does not require predetermined filter or wavelet function. The input images are decomposed into number intrinsic functions (IMFs) as well residual image. performed region by based on...